Вы находитесь на странице: 1из 139

The Economic Impact of

Air Service Liberalization

air transportation

global aviation markets

new travel options

job creation

more service

to more cities
InterVISTAS-ga2 would like to thank the sponsors, listed above, for their
support in this study. The study results further illustrate the benefits to
communities that have liberalized, or plan to liberalize and expand their
air services agreements.
-----TABLE OF CONTENTS

Executive Summary ............................................................................... ES-1


I. Background ........................................................................................ 1
A. Introduction and Objectives ........................................................... 1
II. The Scale of Air Transportation.............................................................. 4
A. Macroeconomics ........................................................................... 4
B. Industry Economics....................................................................... 7
III. The Economic Impact of Liberalization .................................................. 10
A. General - Current Evidence .......................................................... 10
B. Air Service Liberalization and Traffic Growth ................................... 11
C. Air Traffic and Economic Growth ................................................... 14
D. Catalytic Effects from Air Service Development ............................... 16
E. Conclusions ............................................................................... 17
F. Air Service Liberalization - Case Studies ......................................... 18
1. U.S.-U.K. ............................................................................ 19
2. Intra EU.............................................................................. 27
3. UAE-U.K. & Germany ............................................................ 38
4. Trans-Tasman ..................................................................... 44
5. Malaysia-Thailand ................................................................ 52
IV. The Model Description & Methodology................................................ 59
A. Introduction ............................................................................... 59
B. Overall Methodology.................................................................... 59
C. The Cross-Sectional Model for Passenger Traffic .............................. 63
D. Estimation of Passenger Model...................................................... 72
E. Estimation of Cargo Model............................................................ 76
F. Economic and Catalytic Impacts.................................................... 80
G. Summary of Model...................................................................... 89
H. The Model Issues .................................................................... 92
V. Findings and Conclusions.................................................................... 95
Appendix A: The International Aviation Framework ....................................... A-1
Appendix B: Glossary of Terms .................................................................. B-1

i
Executive Summary
The Economic Impact of
Air Service Liberalization
EXECUTIVE SUMMARY

This study found extensive and significant evidence that supports the generally accepted
conventional wisdom that liberalization of air services between countries generates
significant additional opportunities for consumers, shippers, and the numerous direct and
indirect entities and individuals affected by such liberalization. Conversely, it is also
evident that restrictive bilateral air services agreements between countries stifle air travel,
tourism and business, and, consequently, economic growth and job creation.

Prominent findings of the study are:

Traffic growth subsequent to liberalization of air services agreements between


countries typically averaged between 12 percent and 35 percent, significantly greater
than during years preceding liberalization. In a number of situations, growth exceeded
50 percent, and in some cases reached almost 100 percent of the pre-liberalization
rates.

A simulation of the likely results of liberalizing 320 country pair markets that are not
today in an Open Skies (deregulated) mode indicate traffic growth, on average, of
almost 63 percent. This is substantially higher than typical world traffic growth of
around 6 percent-8 percent. Liberalizing only these 320 bilateral agreements of the
2,000 in our database would create 24.1 million full-time jobs and
generate an additional $490 billion in Gross Domestic Product.
Liberalizing only 320
This corresponds to an economy almost the size of Brazil.
bilateral agreements of
The creation of the Single European Aviation Market in 1993 led the existing thousands
to an average annual growth rate in traffic between 1995 and would create 24.1 million
2004 that was almost double the rate of growth in the years 1990 full-time jobs and
to 1994. This produced about 1.4 million new jobs.
generate an additional
A simulation of full liberalization of the United States-United $490 billion in Gross
Kingdom market under a Comprehensive First Step Air Service Domestic Product. This
Agreement (ASA) between the United States and the European corresponds to an
Union would produce an almost 29 percent increase in traffic.
economy almost the size
Some of the increase results from the impact of lower fares, while
of Brazil.
the remainder would result from allowing any U.S. city to obtain
nonstop service to Londons Heathrow or Gatwick airports.

The economic benefits of this liberalization would be substantial. There would be an


additional 117,000 new jobs generated, and the incremental GDP impact to both the
U.S. and to the U.K. would be roughly $7.8 billion.

An examination of 190 countries and 2,000 bilateral air service agreements suggests
that there are still a number of countries that place a priority on protecting their flag
carrier(s), rather than enhancing the overall welfare of the broader public interest.

ES-2
A. BACKGROUND

Commercial aviation owes its existence to the rapid development and application of
technology. Modern aerospace technology allows aircraft to operate efficiently and safely
under a very wide range of conditions, to areas and climates throughout the world. Air
service is widely available, and allows even the poorest nations access to the most
advanced products. The new, ultra-long range aircraft can operate nonstop flights to
points so distant that airlines must decide whether to fly east, over the Atlantic and
northern Europe, or west across the Pacific and the Far East to reach their destination. But
the most important contribution of technology has been to lower the cost of air travel.
Fuel-efficient engines and aerodynamic surfaces, low maintenance and modular
components, and improved materials have progressively allowed airlines to lower air fares,
thereby allowing more and more people to use air transportation on a routine basis. This
diffusion has permitted commercial aviation to play a far more important role in peoples
lives.

However, commercial aviation still faces a challenge common to many of the newer and
more technically advanced areas of our society. Air transportation political and trade
institutions have not kept pace with the evolution of technology or the needs of the public.
Commercial aviation remains encumbered by well meaning but outmoded and arcane
rules, principles and institutions. These rules and regulations often prevent fit, willing and
able airlines from fully serving passengers and shippers who are completely willing and
able to pay. They also impose protective machinery that frustrates innovation, and have in
the past directed the evolution of the industry into a contrived and artificial structure. By
sheltering airlines from market forces, they reduce the incentives for them to pass the
benefits of improved technologies on to passengers, shippers and investors.

International air commerce today is still, in many respects, governed by a framework of


rules laid down in the post World War II era. Despite todays trend toward global markets,
free trade, the internet, and the economic integration of entire continents, one of the most
globalized, technology-intensive industries remains encumbered by rules that stifle
competition and prevent airlines, communities, passengers, and shippers from benefiting
to the fullest. The bilateral air service agreements (ASAs) that continue to govern much
of world trade in aviation define the terms under which airlines will link their two home
territories. These ASAs often frustrate market growth, force users to pay a price premium,
and create a series of vested interests.

The proponents of continuing protection are often large, powerful vested interests, who
consider that they have much to lose from its abolition. The beneficiaries are often small
and fragmented. They include actual and would-be passengers and shippers, hotel
operators, airports, and the huge numbers of people who could or might be employed in
the tourism, transportation, or manufacturing industries. The collective benefits, while
very large, are so widely distributed that few persons or organizations perceive that they
have a major interest in reform. The benefits, permeating throughout the economy, are

ES-3
often so difficult to trace that many are not even aware that they could benefit. Often,
even those who oppose liberalization could gain, but would face certain transitional risks,
and would have to modify their business methods.

This imbalance has fostered a comfortable status quo, in which the wide but diffuse
benefits of air liberalization are subordinated to the interests of the minority. A society
that is unaware of the magnitude of the benefits, whose individuals are unaware that a
change could help them, and cannot estimate the payoffs in any meaningful way, is
unlikely to be able to reap the full rewards of liberalization.

These factors suggest that a key to reforming the regulatory environment surrounding
international air travel is education. Specifically, a society can only make a rational
choice between protectionism and competition if it knows:

That the incremental benefits can be very large;

That the benefits are widely diffused among many individuals and organizations;

That many sectors could benefit, such as the tourism industry, trade/transportation
and manufacturing;

That many persons who may not perceive themselves as actually benefiting could in
fact be made better off;

That even those most opposed to the change could benefit if they can change their
behavior accordingly; and

That these benefits can often be gained at minimal public expenditure.

B. OBJECTIVES OF THE STUDY

The broad objective of this study is to quantify the results of both historical and
prospective bilateral air service agreement (ASA) liberalization. In order to accomplish that
objective, a number of subsidiary programs have been pursued. Thus, InterVISTAS-ga2
examined the economic consequences of the liberalization of air transport throughout the
world. Specific subsidiary objectives were:

To examine recent instances of air service liberalization, or lack thereof, and identify
their most important consequences on competition, traffic growth, carrier behavior and
national economic benefits;

To develop a flexible and robust analytical model, with all associated databases, so that
the benefits of liberalization can be quantified prospectively for any arbitrary
country-pair, or groups of country-pairs; and

ES-4
To promote a more informed public debate on the historical and potential benefits of
liberalization of air services agreements between countries.

This study summarizes the research performed in pursuit of these objectives. This project
has created a mathematical framework of worldwide applicability. While the model is
unique in its sophistication and versatility, the project is still but one of a large group of
analyses of the consequences of liberalization. While the assumptions, methodologies and
results vary widely, the models clearly demonstrate the importance of high quality and
reasonably priced air service to world economic development and job growth.

C. FINDINGS & CONCLUSIONS


This report
This study demonstrates that air service liberalization can promote traffic summarizes the
growth, with an accompanying growth in non-aviation sectors. The sheer development of a
scale of the largest airports, the global reach of the industry, and its methodology to
technological innovation, supports the often cited statistic that the travel
quantify the
and tourism industry drives 12 percent to 15 percent of the world output of
benefits of
goods and services.
international
air service
1. THE ECONOMIC IMPACT OF AIR SERVICES CURRENT EVIDENCE liberalization.
This report summarizes the development of a methodology to quantify the
benefits of international air service liberalization. The approach developed is
unique in that it can apply to any arbitrary country-pair and any level of liberalization.
Extensive research attests to the importance of commercial aviation to nations in all states
of development. Air service liberalization, which replaces a set of strict and arcane rules
with the primacy of the market, has repeatedly proven to be a decisive influence in
expanding the industry and making its benefits available to more people. Many airports,
airlines, academic institutions, governments and private organizations have documented
the relationship between liberalization and economic growth. These efforts have
contributed greatly to our knowledge of liberalization. However, most research has been
narrowly focused in one or a very few specific markets. Most of the work has been ex
poste and retrospective, contrasting a situation before and after liberalization. The data
and models have been very situation-specific, and could not be quickly and simply applied
to other markets.

This study describes a framework to assess the economic benefits of international air
service liberalization in any market, anywhere in the world. Its approach is ex ante; it
estimates the impact of liberalization on any market that is presently restrictive. Its global
applicability depends on the use of data generated throughout the world, involving over
190 nations and 1,400 country-pairs. The various statistical relationships that form the
model do not merely accommodate but, indeed, require this diversity. The model has a

ES-5
wider applicability and greater robustness than those developed from more limited and
homogeneous sets of data.

In keeping with the global focus of the research, the study draws on, to the fullest extent
possible, experience obtained throughout the world. The United States, because of its size,
and the relatively lengthy period since its domestic market was deregulated, offers among
the best examples of market liberalization. Furthermore, American communities and
airports have been most active in pursuing new services, and in evaluating the economic
impacts of aviation.

Section III of this study examines recent evidence on air service liberalization. It does not
purport to provide a detailed review of the literature. Rather, it summarizes the highlights
of recent research, and establishes a point of departure for the approach to developing the
model and related framework.

This study, and most others, is based on a causal chain that links changes in air service
regulation to changes in the broader economy (Figure ES-1).

Figure ES-1: The Causal Relationship between Air Service Liberalization and Economic Growth

New and Job


Traffic Economic
Liberalization better
growth growth growth
air services

The failure of any one link can halt this process of expansion. Sometimes, the current
regulations, however restrictive, do not constrain market behavior. Policy makers may
authorize new services, but if airlines do not wish to operate it, the liberalization would be
irrelevant. Many bilateral agreements are rife with unused authority, services that are
allowed but that have no commercial value. The logical and empirical link between better
air services and traffic growth is much stronger, and all evidence suggests that the market
responds to improved service. The link between traffic growth and economic growth
depends on the level of employment, the propensity to import, and whether the increased
air travel diverts expenditures from other forms of consumption, savings and investment.

2. AIR SERVICE LIBERALIZATION AND TRAFFIC GROWTH


Airlines are continually fine-tuning their routes to accommodate traffic growth, changes in
aircraft technology, airport congestion and other factors. In a mature market, this results
in a never-ending trickle of schedule changes. When the market fundamentals
experience a sudden and dramatic change, a torrent of new routes often results. Such

ES-6
events can include rapid economic growth of the type being experienced in China and
India, the availability of new air routes, and, most importantly, market liberalization.

As of 2006, many nations have allowed market forces to govern their domestic routes,
while a slow, erratic process of creeping liberalization has prevailed on many
international air corridors. Liberalization has promoted many new services around the
world, as testified by Table ES-1.

Table ES-1: Liberalization and Air Service Growth

Event Results
U.S. deregulation, 1978 Emergence of hub and spoke systems, low cost carriers with
nationwide route networks, new entrants and integrated cargo
carriers.
U.K Liberalization of Secondary Airports Growth of international services to Manchester, Birmingham,
Glasgow, etc.
Open Skies Agreements for United Arab Growth of Dubai as major international hub.
Emirates
Domestic deregulation in India Development of low cost carriers and aggressive, expansion-
oriented airlines.
U.K-India Bilateral and Creation of New Growth of capacity, new gateways and additional carriers
Frequencies operating U.K.-India service.
Domestic deregulation in Brazil Growth of low cost carrier Gol and others.
Single European Market Growth of low cost carriers. Ryanair, Easyjet, etc. New services,
traffic growth, new gateways throughout European Union.

Published aviation statistics testify to the ability of new air service to stimulate traffic.
Table ES-2 portrays how new services have stimulated traffic. It compares traffic levels in
the year immediately preceding inauguration of the new service to volumes in the first full
calendar year of operation. Most of the examples result from changes in bilateral air
service agreements, or from specific governmental decisions to relax the restrictive
provisions of current agreements.

ES-7
Table ES-2: New International Services and Traffic Growth

City-Pair Service Liberalization Event Gain


Vancouver-Phoenix America West 1995 1995 Canada-U.S. Bilateral 146.4%
Toronto-Minneapolis Air Canada 1995, 1995 Canada-U.S. Bilateral 55.3%
Northwest
Toronto-New Orleans Air Canada 1998 1995 Canada-U.S. Bilateral 41.2%
Ottawa-Chicago Air Canada/ American 1995 Canada-U.S. Bilateral 109.7%
1995
Montreal-Atlanta Delta 1995 1995 Canada-U.S. Bilateral 55.5%
Atlanta-San Jose CR Delta 1998 1997 U.S.-Costa Rica 118.5%
Dallas/Fort Worth- American 1996 Assignment of routes 336.6%
Santiago
Chicago-Hong Kong United 1996 (not daily) U.S.-Hong Kong Bilateral 21.1%
Chicago-London United 1995 U.S.-U.K Mini Deal, 1995 42.1%
Chicago-Sao Paulo United 1997 U.S.-Brazil, 1996 80.4%
Chicago-Buenos Aires United 1998 Reassignment of routes 41.1%
Houston-Sao Paulo Continental 1999 U.S.-Brazil, 1997 120.5%
Atlanta-Guadalajara Delta 1999 U.S.-Mexico, 1991 169.5%
Washington-Buenos United 2002 Reassignment of routes 208.7%
Aires
Washington-Sao Paulo United 2002 Reassignment of routes 88.4%
Detroit-Beijing Northwest 1996 U.S.-China, 1995 174.3%
Dallas/Fort Worth-Lima American 1996 Assignment of routes 482.0%
Houston-Tokyo Continental 1998 1998 U.S.-Japan 116.6%
Atlanta-Rome Delta 1999 1998 U.S.-Italy 110.8%
Dallas/Fort Worth- American 2000 1995 Open Skies 115.3%
Zurich

Sources: United States Department of Transportation Databases 1B and 28IS; Statistics Canada Report 51-205, Air
Passenger Origin and Destination, Canada-United States Report, and InterVISTAS estimates.

Table ES-2 understates the stimulation of new traffic into a market by using a strict year
before/year after timeframe. Traffic usually requires several years to adjust fully to a new
service. Despite the conservative approach, nonstop international services can often cause
international traffic to double in only a year, even for city-pairs that already have a
profusion of one-stop connecting services. Any mechanism that allows international
services to proliferate to non-traditional gateways can be a powerful stimulus to traffic.
Restrictive bilateral agreements, through confining service to a few named points, can
thwart the growth. They also exacerbate the airside and groundside congestion at the
largest gateways.

ES-8
Figure ES-2: Growth of United States-Canada Traffic, 1990-2004
Passengers (in millions)

20

18

16

14

12
Liberalized Market
10

1990 1992 1994 1996 1998 2000 2002 2004

Source: United States Department of Transportation Database 28IM

Figure ES-2 depicts United States-Canada passenger flows. Until 1995, this market was
governed by a very restrictive bilateral, negotiated in 1966 and updated in 1974. It
prohibited nonstop scheduled services on a large number of routes, including Toronto-
Washington, Atlanta-Montreal and Vancouver-Denver. Notwithstanding that the two
countries had concluded a free trade agreement in 1988, extended to Mexico in 1994;
aviation functioned under a constrained bilateral air services agreement. The liberalization
of air service in 1995 allowed carriers of either nation to serve any route desired, at
commercially determined prices. As shown by the graph, the previously stagnant traffic
saw rapid growth after liberalization.

3. AIR TRAFFIC AND ECONOMIC GROWTH


Tables ES-1 and ES-2 indicate a strong causal relationship between liberalization, air
service improvement, and international traffic. Table ES-3 explores the final step of the
causal chain; the relationship between traffic and economic development. Many airports
have prepared economic impact statements to quantify their influence on their
communities. Several measures are used, including Gross Domestic Product, output,
employment, investment, and tax revenues. Several methods are available, and
assumptions vary widely between each such project. Despite the methodological
differences, the studies have reached a worldwide consensus that airports and civil

ES-9
aviation can have an enormous and positive impact on regional prosperity. Table ES-3
summarizes economic impact statements for a cross-section of airports and civil aviation
activities throughout the world.

Table ES-3: Economic Impact of Commercial Aviation

Passengers Employment Output Source


Des Moines, 1998 1.7 million 2,352 $182 million U.S. Des Moines International Airport
Newcastle, NSW .76 million 3,336 $540 million AU Newcastle Airport Limited
2005
Cincinnati, 2004 22 million 89,536 $5 billion U.S. University of Cincinnati
Reykjavik, 1998 1.8 million 1,156 11.4 Billion IKr University of Iceland Institute of
(2006) Economic Studies
Geneva, 1999 7 million 24,000 9.0 Billion SFr Aroports Internationale
Geneva
World Aviation, 2005 2 billion 29 million $2.96 trillion Air Transport Action Group
U.S.,
8% of world
GDP
U.K. Airports, 2004 229 million 580,000 22.2 billion Airport Operators Association,
gross value 2005
added
Toronto, 2001 28 million 138,000 $14 billion CD Greater Toronto Airports Auth.
Auckland, 2001 8.5 million 235,780 $14.2 billion NZ Auckland International Airport
All United States, 746 million 12.3 million $1.37 trillion U.S. Air Transport Association
2005
Inverness Airport, .5 million 2,297 120 million Inverness and Nairn Enterprise,
2005 Highlands and Islands
Enterprise and Highlands and
Islands Airports Limited
U.K. Air Freight 2,526,266 80,000- 4.99 billion U.K. Air Freight Study Report,
Industry, 2000 tonnes 100,000 U.K. Department of Transport

The instances shown in Table ES-3 describe the impacts of facilities operating with a given
level of service. A further refinement involves measuring the incremental impact of a
specific improvement in air service.

4. CATALYTIC EFFECTS FROM AIR SERVICE DEVELOPMENT


The relationships explored in Table ES-3 view the various impacts as the response of a
pre-existing economy to incremental changes in civil aviation. It assumes that there will
be no changes to the underlying structure of the regional economy. Growing evidence
indicates that new air services can lead to changes in the underlying structure of the
regional economy by creating new capabilities, and forming a different set of transactor
expectations. These reactions can literally create new industries in a region, and allow the
area to compete for economic opportunities throughout the world. These catalytic

ES-10
impacts are the most difficult to quantify. Although most evidence is anecdotal, there is
growing evidence that these effects can be large:

A 10 percent increase in the supply of intercontinental flights creates around a


4 percent increase in the number of headquarters of large firms located in the
corresponding urban area headquarters of knowledge-intensive sectors are much
more influenced by the supply of direct intercontinental flights than are those of non-
knowledge-intensive sectors.1

Nine foreign-owned companies in Northern Kentucky cited air service as an important


factor in their choice of location. The nine firms collectively employ 1,470 persons.2

In 2003, Kenya exported 50,000 tonnes of flowers by air freight.3

The growth of European air transport since 1995 has boosted European Union GDP by
4 percent. The expected growth to 2025 will boost GDP of the 25 European Union
nations by a further 1.8 percent.4

Air service liberalization in Egypt could increase the GDP of the Travel and Tourism
industry by 12 percent by 2011, adding 260,000 full time jobs. Furthermore, the total
GDP for all sectors would increase by 1.8 percent.5

The evidence cited demonstrates a strong causal link between air service liberalization,
traffic growth and economic development. It lends further strength to the need for air
service liberalization by illustrating the benefits of successful efforts, and the harmful
consequences that have resulted from retaining the status quo.

5. THE ENVIRONMENT
Commercial aviation industries are driven by the business realities of competitive market
forces that drive continuous innovations in technology that increase airplane efficiencies
and minimize environmental impacts. The aviation industry also recognizes that its
members must share the leadership required to minimize environmental impacts resulting
from air travel growth.

1
Germ Bel and Xavier Fegeda, Getting There Fast: Globalization, Intercontinental Flights and Location of
Headquarters, Unpublished, September 2005.
2
George Vredeveld, Marie Haney, Pooja Sharma and Anthony Apostolides, The Influence of International Airports
on Regional Economic Growth. Economics Research Group, Center for Economic Education, University of Cincinnati,
Cincinnati, 1997.
3
Oladele Samson Fatokun, African Air Transport in the 21st Century: A Case Study of the Contrasting Experience of
Nigeria and Kenya, MSc Thesis, Department of Engineering, Cranfield University, 2005.
4
The Economic Catalytic Effects of Air Transport in Europe, Eurocontrol Experimental Centre, Bretigny sur Orge
Cedex, 2005.
5
Aviation in Egypt: The Impact on Travel & Tourism, Jobs and the Economy, World Travel and Tourism Council,
2005.

ES-11
Industry Environmental Initiatives Some of the current industry initiatives designed to
minimize environmental impacts include:

The industry is improving fuel efficiency and minimizing emissions

> fuel efficiency and associated emission reductions have improved by more than 70
percent over the last 40 years, and aircraft generate only between 2 percent and 4
percent of global CO2 emissions
> manufacturers continue improvements in aerodynamics and reductions in weight,
including new technologies such as winglets, raked wing tips, and composite
materials
> flying on todays new airplanes is one of the most fuel efficient ways to travel the
newest airplanes are as or more fuel efficient than the average car
> the industry is pursuing conservation technologies and alternative fuels, operating
research and technology centers and contributing technical patents to research
institutions

The industry is minimizing noise

> manufacturers are making quieter airplanes the newest airplanes have achieved
noise footprints that fit within the airport boundaries of most airports
> industry partnerships are developing and implementing quiet arrival and departure
techniques, including the continuous descent approach that will achieve noise
reductions by as much as 35 percent

The industry is achieving business efficiencies that minimize environmental impacts

> manufacturer fleets deliver new airplanes for every market to minimize empty seats
and are cleaner, quieter and more efficient
> new airplanes are creating greater opportunities for point-to-point travel that reduce
takeoffs and landings and typically save fuel over the hub-and-spoke, connecting
flight approach

The industry is promoting efficient air traffic management (ATM), recognizing that ATM
efficiency benefits represent the greatest short-term opportunities for minimizing
emissions and noise impacts. Aircraft manufacturers, airlines, regulators and airports
are working together in partnerships to increase capacity and at the same time
improve environmental performance through:

> Priority departures


> Continuous descent

The industry is addressing global aviation environmental impacts in systemic ways

ES-12
> the industry is using life cycle management approaches to environmental
considerations that incorporate opportunities to minimize environmental impacts of
aviation growth and aircraft development over the entire life cycle from
research/development through retirement

> the industry recognizes that all components of transportation systems need to be
included when considering the environmental impacts of transportation vehicles,
including:
infrastructures roads, railways, airports, etc.
energy sources fuel production processes/facilities, electrical production
processes/plants, etc.

On balance, there are a number of system and technology driven initiatives that are being
developed and implemented by the aircraft manufacturer and commercial airline industries
to minimize environment impacts through fuel efficiency and associated emissions
reductions, and minimizing noise.

D. AIR SERVICE LIBERALIZATION CASE STUDIES

1. GENERAL RESULTS
In order to test the hypothesis that liberalization leads to market growth and economic
expansion, and to validate the results being generated by the economic model, we studied
five separate cases.

In each case, we examined the


US-UK Germany/UK-UAE
background of the bilateral
relationship, the history of traffic
growth, and its relationship to
benchmark parameters such as
GDP growth. In all cases
studied, it was apparent that,
depending on the size and
development of the economies,
there was substantial
Intra EU
incremental passenger traffic
and economic growth after air
Malaysia-Thailand
service agreements between the
countries had been liberalized.
In some cases, the liberalization
was of a transitional nature -
that is, from a rigid Bermuda I
type agreement to something
Trans-Tasman

ES-13
less than Open Skies, while in other cases the liberalization was from a transitional to an
Open Skies regime. In one case, we found that liberalization occurred as a result of
informal understandings between governments, with no accompanying modifications to
the formal air services agreement.

Post liberalization traffic growth tended to exceed pre-liberalization growth levels by


anywhere between 12 percent and 35 percent and up to 50 percent, depending on the
periods measured. In all cases, the traffic growth produced significant increases in
economic output and job growth.

A short summary of the individual cases is outlined below.

2. SPECIFIC CASE STUDIES

United States-United Kingdom

In 1995, the restrictive Bermuda II agreement saw a partial easing. Airlines of both the
United States and the United Kingdom obtained unlimited access between any set of
airports, with the conspicuous exceptions of London Heathrow and Gatwick. United Airlines
obtained Chicago-London rights.

These steps caused a steady expansion of air services and traffic. Since 1995, traffic
between Chicago and London has more than doubled. Services have expanded at
Manchester, Birmingham and Glasgow, while Bristol and Edinburgh have emerged as
trans-Atlantic gateways.

The economic benefits have been significant. By 2004, the additional traffic and services
generated 9,197 full-time jobs in the United States and over 16,700 in the United
Kingdom. The Gross Domestic Product of the United States expanded by $747 million; the
United Kingdom saw a $970 million increase.

May 1994

Nonstop US-UK Routes (Excluding Heathrow & Gatwick)


Source: May 1994 Official Airline Guide, US/UK Designated Carriers Only

Boston

New York JFK


Chicago OHare

Glasgow
Washington Dulles
Belfast
Los Angeles

Atlanta Manchester

Birmingham

ES-14
May 2006

Nonstop US-UK Routes (Excluding Heathrow & Gatwick)


Source: May 2006 Official Airline Guide, US/UK Designated Carriers Only

Boston

New York JFK


Newark
Chicago OHare Philadelphia Glasgow
Washington Dulles
Edinburgh

Las Vegas
Belfast
Atlanta
Manchester

Birmingham
Orlando (MCO)
London Stansted
Bristol

With respect to the future, should there be a Comprehensive First Step Air Transport
Agreement between the U.S. and the EU, we can expect the impact on the economies of
the U.S. and the U.K. to be significant.

A simulation of full liberalization of the United States-United Kingdom market as a result of


a Comprehensive First Step Air Transport Agreement between the U.S. and the European
Union would result in a 29 percent increase in traffic. The increase would derive in part
from lower fares, and in part from allowing any U.S. city to obtain nonstop service to
Londons Heathrow or Gatwick airports.

The economic benefits of this liberalization would be substantial. Over 117,000 new jobs
would be created, and incremental GDP would approximate $7.8 billion.

Intra European Community

The liberalizations that created the Single European Aviation Market dramatically increased
intra-European air travel.

The 1992 package that created the Single European Aviation Market did away with
bilaterals for services within the Community. Its main provisions were:

Community air carriers were permitted to exercise traffic rights on routes anywhere
within the Community. Until 1997 this included only en route cabotage (e.g. Air France
flying Paris-Frankfurt-Berlin) provided that no more than 50 percent of the capacity
was used for the cabotage service. Freestanding cabotage was liberalized from 1997.

Safeguards were provided to protect routes where a public service obligation existed,
particularly thin routes operated by small aircraft. There was also a provision to enable

ES-15
Member State A to refuse access to a service using an airport in Member State B to
which the airlines of Member State A could not get access for reasons of congestion. It
was also possible to refuse or limit the use of traffic rights where serious congestion or
environmental problems existed. None of these safeguards has been extensively used
and they can be ignored for the purpose of this exercise.

Capacity limitations were made illegal except in cases of congestion or environmental


problems.

Community carriers were free to set airfares. Charter fares and seat and cargo rates
charged by Community air carriers were to be set by free agreement of the parties to
the contract of carriage.

Safeguards were provided to deal with excessively low or excessively high fares. They
have never been used.

A significant impact of the 1992 package was the stimulus it gave to the development of
low-cost airline services. An operator could order aircraft confident in the knowledge that
he had access to a large market without legislative restrictions was (and is) a major
encouragement to investment in new services and in the aircraft necessary to carry them
out. The table below sets out the low-cost operators share of capacity from 1996 to 2003.
Their impact on the market did not begin until 1996. Until about 2000, their presence was
concentrated on routes from the U.K.

Low-Cost Operators
Year 6
Share of Capacity
1996 1.4%
1997 2.8%
1998 3.7%
1999 4.2%
2000 6.0%
2001 6.4%
2002 11.1%
2003 20.2%

The Single European Aviation Market resulted in the generation of an incremental 44


million passengers, an increase in post-liberalization years of over 33 percent as
contrasted with historical Intra-European market growth of between 4 percent and 6
percent per annum. The additional traffic required an additional 681 daily return flights.

6
Percentage of weekly available seats provided by low-cost operators between 1996 and 2003; Source: European
Commission (2005). DG TREN: Analysis of the European Air Transport Industry 2003 (Contract number B2-7040B-
S07.17962). Brussels: Office for the Official Publications of the European Union.

ES-16
The traffic expansion spurred development of both the travel/tourism sector and other
industries. Fully 1.4 million full-time jobs resulted from the liberalization, and the
European GDP grew by $US 85 billion (62 billion Euro).

United Arab Emirates to the United Kingdom and Germany

Liberalization of the Germany-UAE market in 1986 resulted in high growth rates (both in
terms of ASKs and passengers carried), although the absolute traffic figures are not large.

Although the U.K.-UAE route was liberalized much later, the commercial situation on the
two routes in 1998 was similar. There were only nine weekly flights by Emirates/Gulf Air
between Germany and the UAE. On U.K.-UAE routes, British Airways ran a double daily
service and Emirates/Gulf Air a similar one. But in 1998, Emirates was beginning its
expansion; in the case of the U.K., Virgin Atlantic was pressing for Gulf route authority.

The conclusion we draw from this analysis is that the major impact of the liberalizing of
the two bilaterals lies in the scope they gave to the development of Emirates and in
particular to the development of Dubai as a hub for traffic between Europe, on one hand,
and the Far East and Australasia on the other. Without Emirates expansion, we would
expect the growth rates to have increased as a result of the development of Dubai as a
financial, trading and tourist center, but not to the same extent. Thus, it is reasonable to
conclude that the liberal bilateral arrangements between the UAE and the two major
financial centers of the EU contributed to some degree to the confidence which is
necessary for such developments.

Germany

In 2004, over 1 million persons traveled between Germany and the United Arab
Emirates. The liberalization of 1986, allowing a proliferation of direct services between
the two countries, prompted a market expansion of almost 167,000 passengers. Traffic
was fully 19.7 percent higher than it would have been in the absence of liberalization.

The expanded traffic benefited both nations. In the United Arab Emirates, the
equivalent of 745 new full-time positions was created, with a $15 million increase in
GDP. The 2,600 new full-time positions in Germany accompanied a $152 million growth
in GDP.

United Kingdom

The 1998 liberalization, combined with a rising price of oil and rapid expansion by
Emirates Air and later, Etihad, prompted a 59 percent increase in total United
Kingdom-United Arab Emirates traffic. By 2005, total country-pair traffic exceeded 3.2
million passengers, of which 1.1 million could be attributed to the more liberal air
service agreement.

ES-17
Both nations benefited significantly. In the United Arab Emirates, over 5,300 persons
found full-time positions, and GDP expanded by over $110 million. The expansion created
over 18,700 positions in the United Kingdom, and over one billion dollars additional GDP.

Malaysia-Thailand

The original agreement between the governments of Malaysia and Thailand was signed in
1969. This type of agreement (Bermuda I) still defines many bilateral relationships around
the world today. While not exactly like the American style of Open Skies, the agreements
Bermuda I style does allow for new services to be introduced to the market. Subsequent
agreed upon Memorandum of Understandings (MOUs) between Malaysia and Thailand has
allowed services and frequencies to increase, thus driving an increase in overall traffic.

When analyzing the current aviation relationship between the two countries, many aspects
of a liberalized market can be found within the rights of the agreement (September 2001
was the last time it was updated with open capacity). This includes no restrictions on
points served in the market, multiple designations allowed on routes, code sharing rights
and open frequencies. Fifth freedom, intermediate or beyond, rights in addition to seventh
freedom cargo operations are not included in the agreement. As in many other bilateral
agreements found throughout the world, cabotage is not included.

Malaysia and Thailand have numerous air service agreements with third countries. They
have both concluded Open Skies agreements with the United States and other major
trading partners. Both currently support a multiple designation policy as a way of
increasing their countries respective role in global tourism and trade. With a new
international airport in Kuala Lumpur and one planned to open soon in Bangkok, a liberal
designation policy helps remove barriers for new carriers seeking access to major
international gateways. With regards to points served, Malaysia and Thailand have set up
open policies not only in their Open Skies agreements, but in other bilateral agreements.
The key restriction on the Malaysia-Thailand bilateral agreement is beyond rights and
seventh freedom cargo operations. Code sharing rights have allowed Malaysia and Thai
Airways to cooperate on key routes between the two countries. In addition, both countries
heavily promote secondary international destination access rights for tourism (Phuket,
Thailand Langkawi, Malaysia Penang, Malaysia).

In 2005, 1.3 million passengers traveled between Thailand and Malaysia. Of this total,
over 370,000 can be attributed to the combination of the liberalized regime and the entry
of low cost carriers. This suggests that the direct and indirect effects of liberalization have
caused a market expansion of over 37 percent.

The economic and tourism impacts of this increased traffic demand on Thailand and
Malaysia are identical. Each nation obtained more than 4,300 full-time positions and a
stimulus of over $114 million to GDP.

ES-18
Australia-New Zealand (Trans-Tasman)

The first steps towards economic liberalization between these neighbors can be traced to
1966, when the New Zealand and Australia Free Trade Agreement was signed. This
agreement was in place for 17 years until March 28, 1983, when the Australia-New
Zealand Closer Economic Relations Free Trade Agreement (ANZCERTA) was concluded.
The ANZCERTA set a foundation as an innovative agreement, which not only created a
liberal business and economic regime for goods and services, but also set a collaborative
umbrella to deal with customs, transport, regulatory, product standards and business law
issues. The ANZCERTA established a market that continues one of the most open
economic trade relationships between any two countries in the world. The ANZCERTA is
continually reviewed to ensure that the agreement remains effective in all sectors of the
economy.

This analysis focuses specifically on the effects of the Single Aviation Market that was
established in 1996, preceding full Open Skies in 2000.

Single Aviation Market

Australia and New Zealand concluded a Single Aviation Market (SAM) agreement, effective
as of November 1, 1996. The goal of the Single Aviation Market was to bring the two
countries closer together within the elements of the ANZCERTA. The main components of
the agreement included the opening of ownership and control regulations in the bilateral
market, the introduction of unlimited frequencies for Trans-Tasman services and a
provision that allowed airlines of either country to operate domestic flights within the other
country. While the SAM agreement opened up many new opportunities within the Trans-
Tasman market, it did not address beyond markets to third countries. Those markets were
still under the original 1961 Australia New Zealand Air Services Agreement and the
subsequent 1992 Memorandum of Understanding. Two different definitions of air carriers
were created from the agreement: the Domestic and the SAM airline. The Domestic
airline designation allowed carriers to fly domestic services in each others domestic market
and the SAM designation harmonized ownership, control, technical and safety certifications
from each countries regulatory agencies.

The importance of the Single Aviation Market agreement was that it broke barriers in the
carriage of cabotage traffic, created ownership and control flexibilities, and deregulated
capacity, designations, and frequencies. More importantly, the SAM agreement established
the foundation for a more liberal agreement that, in the future, would open markets
beyond the Trans-Tasman.

The liberalization of 1995 spurred a rapid growth in traffic between Australia and New
Zealand. Other factors, including the entry and exit of domestic airlines in both nations,
and changes resulting from fifth freedom activity, confound any effort to measure the

ES-19
distinct impacts of the relaxation of third/fourth freedom restrictions. However, these
events themselves are the direct or indirect results of other liberalization efforts.

By 2005, Australia-New Zealand traffic was fully 56 percent higher than it would have
been in the absence of any liberalization. The relaxed market controls increased total
traffic by over 1.7 million passengers per year. The additional volume would require a
further 27 flights daily.

Each nation gained more than 20,600 full-time positions from the liberalization and the
ensuing traffic increase. The GDP of each country grew by $726 million U.S.

ES-20
I. Background
The Economic Impact of
Air Service Liberalization
I. BACKGROUND

A. INTRODUCTION AND OBJECTIVES

Flying has always been a symbol of freedom. Only in the last hundred years has the dream
of flight been realized, and only in the most recent handful of decades has it become
widely available. Today, widespread access to flight gives humans the elemental freedom
to soar that they have sought for centuries. More importantly, it grants us the freedom to
travel, to see new places, encounter new ideas, challenge outmoded stereotypes, obtain
access to unique goods, and develop our lives on a global scale. The benefits of
commercial air travel are not confined to its immediate users. Entire industries depend on
it to bring in tourists, to create access to distant markets for high value or perishable
products, and to allow businesses and individuals to look to far off regions for their
livelihoods and growth opportunities.

Commercial aviation owes its existence to the rapid development and application of
technology. Modern aerospace technology allows aircraft to operate efficiently and safely
under a very wide range of conditions, to areas and climates throughout the world. It is
widely available, and allows even the poorest nations access to the most advanced
products. The new, ultra-long-range aircraft can operate nonstop flights to points so
distant that airlines must decide whether to fly east, over the Atlantic and northern
Europe, or west across the Pacific and the Far East to reach their destination. But the most
important contribution of technology has been to lower the cost of air travel. Fuel-efficient
engines and aerodynamic surfaces, low maintenance and modular components, and
improved materials have progressively allowed airlines to lower their fares, thereby
allowing more and more people to use air transportation on a routine basis.

However, commercial aviation still faces a challenge common to many of the newer and
technically advanced areas of our society. Our social and political institutions have not
kept pace with the evolution of technology or the needs of the public. Commercial aviation
remains encumbered by well-meaning but outmoded and arcane rules, principles and
institutions. They often prevent fit, willing and able airlines from fully serving passengers
and shippers who are completely willing and able to pay. They also impose protective
machinery that frustrates innovation and directs the evolution of the industry into a
contrived and artificial structure. By sheltering airlines from market forces, they reduce
the incentives to pass on to passengers, shippers and investors the benefits of improved
technologies.

Air commerce today is still governed by a framework of rules laid down in the first half of
the last century. Despite todays growing internationalism of free trade, global capital
markets, the Internet, and the economic integration of entire continents, one of the most
globalized, technology-intensive industries remains fettered by rules that stifle competition
and prevent airlines, communities, passengers, and shippers from benefiting to the fullest.

1
The bilateral air service agreements (ASAs) that two countries sign to define the terms
under which airlines will link their two territories often frustrate market growth, force
users to pay a premium, and create vested interests that could oppose any reform.

The collective benefits of liberalization, while very large, are so widely distributed that no
person or organization perceives that it has a major interest in reform. The benefits,
permeating throughout the economy are often so difficult to trace that many are not even
aware that they could benefit. Often, even those who oppose liberalization could gain, but
would face certain transitional risks and might have to modify their methods of business.

These factors suggest that the key to reforming air travel is education. Specifically, a
society can only make a rational choice between protectionism and competition if it
knows:

That the incremental benefits can be very large;

That the benefits are widely diffused among many individuals and organizations;

That many sectors could benefit, such as the tourism industry, trade/transportation
and manufacturing;

That many persons who may not perceive themselves as actually benefiting could in
fact be made better off;

That even those most opposed to the change could benefit if they can change their
behavior accordingly; and

That these benefits can often be gained at minimal public expenditure, and in an
environmentally sensitive way.

In light of the existing environment, this study was undertaken with the primary
objectives to define and quantify the results of both historical and prospective
liberalization of ASAs. In order to accomplish this objective, a number of subsidiary
objectives have been pursued as well, including:

To examine recent instances of air service liberalization and identify their most
important consequences on competition, traffic growth, carrier behavior and national
economic benefits;

To develop a flexible and rigorous analytical framework, with all associated databases,
so that the benefits of liberalization could be quantified for any arbitrary country-
pair; and

To promote informed public debate on the historical and potential benefits of


liberalization of air service agreements between countries.

2
This study summarizes the research performed in pursuit of these objectives. This project
has created a mathematical framework of worldwide applicability. While the model is
unique in its sophistication and versatility, the project is still but one of a large group of
analyses of the consequences of liberalization. Section III examines other approaches and
their findings. While the assumptions, methodologies and results vary widely, the models
clearly demonstrate the importance of high quality and reasonably priced air service to
world economic development and job growth. They are equally effective in highlighting the
costs of adhering to outmoded and timid practices that stifle growth on an international
scale.

3
The Economic Impact of
Air Service Liberalization

Air Transportation
II. The Scale of
II. THE SCALE OF AIR TRANSPORTATION

A. MACROECONOMICS

Air transportation has long been the life-blood of trade, commerce, and tourism. It plays a
vital role in the economies of individual nations and of the world, and provides
employment for large portions of the world labor force, both directly and indirectly. The
often referred to globalization of world trade can be attributed to many sectors of the
world economy, but few play a more significant role than air transport. As aircraft become
more technologically sophisticated, and as the range of flight is extended to distances
heretofore unheard of, societies, economies, and peoples become more closely bound
together. Moreover, the technological advances in aircraft have also brought improving
economics to commercial air transport, thus opening air travel to more and more of the
world population.

Air transport is the glue that links the world and drives growth and development, while
improving the social welfare of the population. Commercial aviation creates demand in
many sectors of the world economy. The tourism industry is heavily dependent on
aviation, and benefits substantially from the fact that airlines of the world are able to
transport people to tourist destinations at increasingly more favorable prices.

In 2004, Visit Britain estimated that the tourism sector employed around 1.4 million
people in the U.K. or around 5 percent of total employment. Last year around 28
million overseas visitors came to the U.K. spending around 13 billion pounds. Over 70
percent of these visitors arrived by air and they accounted for around 85 percent of
tourism expenditures.7

The globalization of world economies is also dependent on an increasingly efficient and


readily available international air transport system.

Within this context, the quality of passenger transportation networks is a key input
for processing and transmitting information efficiently because it influences the costs
and opportunities for face-to-face contacts between cities. Indeed, it can be argued
that large (global) firms should demand international accessibility when choosing a
headquarters location.8

The worlds airlines transported roughly 2 billion passengers in 2005, and transported
almost 40 percent of world trade (measured by value). The airline industry sector drove
economic expansion, as shown by Figure II-1.

7
AOA speaking for airports. The Economic and Social Impact of Airports, September 2005.
8
Gerrma Bel and Exavier Fegeda,Getting There Fast: Globalization, Intercontinental Flights and Location of
Headquarters, Unpublished. September 2005.

4
Figure II-1: Air Transport The Driver

Commercial Aviation

Tourism Development Aircraft Manufacturing Suppliers, etc.

Economic Impacts

Direct Indirect Induced Catalytic

First, the direct impact of air transport is evident in industry employment and the
development and manufacturing of aircraft and engines, the movement of passengers and
freight, and the direct employment and materials necessary to support the aviation
network. This includes airline employees, air traffic control systems, airports and on-
ground services that are essential to the movement of aircraft, e.g., fueling systems.

The worlds airlines directly employ over 2 million people.

Second, the indirect impact of air transport can be seen in the development of hotel and
tourism industries and the producers and suppliers of the goods and services to the airline
and related industries, e.g. accounting and legal firms, food and beverage suppliers, etc.

Third, induced impacts are driven by, among other things, the spend of the people
that provide support to the industry.

Finally, there are the catalytic impacts of air transport. This includes, for example, the
investment made by organizations in plants and facilities, and the increased trade flows
driven by the expanding capacity of the air transport system to support these
investments.

5
During the late 1970s and early 1980s Atlanta became the focus of a number of
European airlines. As a result of service by Delta from Atlanta to Europe, and
European carriers to Atlanta, foreign direct investment increased dramatically.
For example, there were 67 U.K. companies operating in Atlanta in 1978 and 201
in 1988. Equally dramatic, there were 23 Netherlands companies operating in
Atlanta in 1978 and 116 in 1988, subsequent to the implementation of
international air service.9

The combined economic impact of the industry has been variously estimated at between 4
percent and 8 percent of world GDP. In its recent Annual Report, the International Air
Transport Association (IATA) noted that air transport directly employs four million people
and generates $400 billion in output. In addition, IATA noted that total world output
generated by air transport was nearly $1.4 trillion, 4.5 percent of global GNP.10

Air transport is a major contributor to job creation and economic


growth. It has an economic impact exceeding $2.9 trillion or 8% of global GDP
and generates 29 million jobs globally, falling into the following categories:

Around 5 million are directly employed;


There are 5.8 million indirect jobs
2.7 million are induced jobs
Another 6.7 million direct and 8.8 million indirect jobs are the result of air
transports catalytic impact on tourism 11

The Air Transportation Action Groups study, The Economic & Social Benefits of Air
Transport, also suggests a contribution to world GDP of 8 percent.12

Further cases illustrate the power of air transport to generate economic benefit:

1. European airport studies estimate that European airports contribute 2.6 percent of
European GDP.

2. Air transport impact on the U.S. economy is estimated to generate 12.3 million
jobs and $1.37 trillion in economic impacts.

9
Atlanta Chamber of Commerce.
10
Annual Report, 2005 International Air Transport Association.
11
ComMark Trust The Economic Benefits of Liberalising Regional Air Transport A Review of Global Experience,
November 2005.
12
Air Transport Action Group (ATAG), The Economic & Social Benefits of Air Transport. Pg. 2.

6
3. The U.K. Airfreight Industry is estimated to have created 80-100,000 full-time
equivalent jobs, and almost 5 billion pounds GDP growth13.

While creating consistent growth in world economies, air transport is also increasingly
sensitive to the environment. In this regard, Aircraft entering todays fleets are 70
percent more fuel-efficient than they were 40 years ago. Carbon monoxide emissions have
been simultaneously reduced by 50 percent, while unburned hydrocarbon and smoke have
been cut by 90 percent.14

B. INDUSTRY ECONOMICS

Commercial air transportation worldwide generates $400 billion in revenues. However,


over the past five years, it has managed to lose $40 billion, much of this in the U.S. As a
result, there have been bankruptcy reorganizations in the United States and outright
liquidations of air carriers in other countries where bankruptcy laws do not facilitate such
reorganization.

The world airline industry faced the trauma generated by the tragedy of 9/11, the ensuing
slowdown in the world economies, SARS, the Gulf War, the aggressive expansion of the
Low Cost Carrier (LCC) worldwide, and fuel price escalation at a rate beyond what it could
absorb or pass on to the consumer.

In 2004, the air transport industry marked its fourth consecutive year of financial
aggregate loss, bringing post-9/11 industry losses to over US$36 billion The principal
cause of the $4.8 billion loss in 2004 was the industrys $61 billion fuel bill, $17 billion
more than in 2003. And the situation is getting worse.15

13
Source: United Kingdom Department of Transport, United Kingdom Air Freight Study, 2000.
14
ibid.
15
Annual Report, 2005 International Air Transport Association.

7
Figure II-2: Airline Profit/Loss (2001-2004)
USA and World
$0

$2

$4

Annual Net
Income $6
($billions)

$8

$10

$12
2001 2002 2003 2004
Year

USA
World

Source: Air Transport World, World Airline Report

As of the spring of 2006, however, there are encouraging signs that the industry is moving
toward recovery and expansion. During the first quarter of 2006, average fares increased
throughout the world, although at a more accelerated rate in the U.S. The increases in
airfares are accompanied by cost efficiencies arising from the carrier restructuring that has
taken place over the past four years. The net effect is that the year 2006 is likely to be the
turnaround point in industry earnings.

The cost efficiencies being implemented by carriers around the world are facilitating a
renewed level of competition between and among Low Cost Carriers (LCCs) and between
the LCCs and Legacy Network Carriers (LNCs).

According to the Chairman of American Airlines, Gerard Arpey, the airline is looking
under every rock for cost savings. He noted further that American has targeted $700
million in additional cost savings for 2006.16

And from United Airlines Glenn Tilton, referring to the past three years,

16
Comments of Gerard Arpey at the American Airlines Annual Stockholders Meeting, May 18, 2006.

8
required a discipline and a rigor that will now apply to our operations and our
financial performance, across every aspect of the organization, as we take the
company to the next level.17

This new discipline, which is evident worldwide, will lead to a more competitive and
profitable environment. It will also result in efficiencies that can be translated to
stabilization of air fares over the longer run. It will boost customer demand, particularly in
growing world markets that have not yet reached the scale or penetration of the U.S.
market.

The aircraft market is also recovering. In 2005, Boeing and Airbus booked over 2,100
orders for new aircraft.18 While 2006 orders are forecast by some experts at only around
1,000, this would still reflect a reasonable two-year cycle.

The commercial airline industry continues to face formidable challenges. The industry has
faced up to and resolved many of its inherent cost inefficiencies. For the LNCs, these
uncontrollable adverse factors were compounded by the LCC, which continued to grow in
size, and consequently increasingly dictate fare levels.

The airlines have now taken steps, often through bankruptcy or alternative restructuring,
to bring their costs into competitive balance. Thus, having dealt with the cost issues, the
airlines must now continue to manage their capacity and pricing. Should they have the
discipline to do so, we could see better industry financial health over the next few years,
while maintaining solid traffic growth.

However, the world airline industry is still plagued by the arcane network of bilateral air
service agreements (ASAs). These agreements, negotiated between governments, often
severely restrict the ability of airlines to make sensible and reasonable commercial
decisions relative to where they will allocate their aircraft and at what price to the
consumer. Thus, the consumer, that supposed beneficiary of this regulatory regime, is
really the casualty of regulation. Regulation has proven to be a way to protect the
inefficient, create artificially higher airfares, and stifle growth.

The next section of this study will demonstrate how effectively markets can function when
freed from their historic regulatory bindings.

17
Chicago Tribune, May 8, 2006.
18
The Airline Monitor, May 2006, pgs. 4 & 5.

9
The Economic Impact of
Air Service Liberalization

III. The Impacts of


Liberalization
III. THE ECONOMIC IMPACT OF LIBERALIZATION

A. GENERAL CURRENT EVIDENCE

Extensive research attests to the importance of commercial aviation to nations in all states
of development. Air service liberalization, which replaces a set of strict and arcane rules
with the primacy of the market, has repeatedly proven a decisive influence in expanding
the industry, and making its benefits available to more people. Many airports, airlines,
academic institutions, governments and private organizations have documented the
relationship between liberalization and economic growth. These efforts have contributed
greatly to our knowledge of liberalization. However, most research has been narrowly
focused in one or a very few specific markets. Most of the work has been ex poste and
retrospective; contrasting a situation before and after liberalization. The data and models
have been very situation-specific, and could not be quickly and simply applied to other
markets.

This study establishes a framework to assess the economic benefits of international air
service liberalization in any market, anywhere in the world. Its approach is ex ante; it
estimates the impact of liberalization to any market that is presently restrictive. Its global
applicability depends on the use of data generated throughout the world, involving over
190 nations and 1,400 country-pairs. The various statistical relationships that form the
model not merely accommodate, but, indeed, require this diversity. The models have a
wider applicability, greater robustness, and better statistical properties, than those
developed from a more limited and homogeneous set of data.

In keeping with the worldwide focus of the research, this chapter draws, to the fullest
extent possible, on experience obtained throughout the world. The United States has high
quality statistics on air travel. Because of its size, and the relatively lengthy period since it
deregulated its domestic market (airline deregulation became effective in 1978), the
United States offers the best examples of market liberalization. Furthermore, American
airports have been most active in pursuing new services, and in evaluating the economic
impacts of aviation. This combination provides the best data for researching the subject.

This section first examines recent evidence on air service liberalization, and second uses
case studies to test the hypothesis that liberalization expands the market and adds
economic value and jobs.

This study, and most others, is based on a causal chain that links changes in air service
regulation to changes in the broader economy (Figure III-1).

10
Figure III-1: The Causal Relationship between Air Service Liberalization and Economic Growth

New and Job


Traffic Economic
Liberalization better
growth growth growth
air services

The failure of any one link can halt the process of expansion. Sometimes, the current
regulations, however restrictive, may not be a constraint to market behavior. Policy
makers may authorize a new service, but the tangible benefits of liberalization will appear
only if the airlines exercise the rights. Many bilateral agreements are rife with unused
authority; services that are allowed but that have no commercial value. The logical and
empirical link between better air services and traffic growth is strong, and all evidence
suggests that the market responds to improved service. The link between traffic growth
and economic growth depends on the countrys level of employment, the propensity to
import, and whether increased air travel diverts expenditures from other forms of
consumption, savings and investment.

B. AIR SERVICE LIBERALIZATION AND TRAFFIC GROWTH

Airlines are continually fine-tuning their routes to accommodate traffic growth, changes in
aircraft technology, airport congestion and other factors. In a mature market, this results
in a never-ending trickle of schedule changes. When the market fundamentals
experience a sudden and dramatic change, a torrent of new routes often results. Such
events can include rapid economic growth of the type being experienced in China and
India, the availability of new air routes, and, most importantly, market liberalization.

As of 2006, many nations have allowed market forces to govern their domestic routes,
while a slow, erratic process of creeping liberalization has prevailed on many
international air corridors. Liberalization has promoted many new services around the
world, as testified by Table III-1.

11
Table III-1: Liberalization and Air Service Growth

Event Results
U.S. deregulation, 1978 Emergence of hub and spoke systems, low cost carriers with
nationwide route networks, new entrants and integrated cargo
carriers.
U.K Liberalization of Secondary Airports Growth of international services to Manchester, Birmingham,
Glasgow, etc.
Open Skies Agreements for United Arab Growth of Dubai as major international hub.
Emirates
Domestic deregulation in India Development of low cost carriers and aggressive, expansion-
oriented airlines.
U.K-India Bilateral and Creation of New Growth of capacity, new gateways and additional carriers
Frequencies operating U.K.-India service.
Domestic deregulation in Brazil Growth of low cost carrier Gol and others.
Single European Market Growth of low cost carriers. Ryanair, Easyjet, etc. New services,
traffic growth, new gateways throughout European Union.

Published aviation statistics testify to the ability of new air service to stimulate traffic.
Table III-2 portrays how new services have stimulated traffic. It compares traffic levels in
the year immediately preceding inauguration of the new service, to volumes in the first full
calendar year of operation. Most of the examples result from changes in bilateral air
service agreements, or from specific governmental decisions of how to exercise the
provisions of current restrictive agreements.

12
Table III-2: New International Services and Traffic Growth

City-Pair Service Liberalization Event Gain


Vancouver-Phoenix America West 1995 1995 Canada-U.S. Bilateral 146.4%
Toronto-Minneapolis Air Canada 1995, 1995 Canada-U.S. Bilateral 55.3%
Northwest
Toronto-New Orleans Air Canada 1998 1995 Canada-U.S. Bilateral 41.2%
Ottawa-Chicago Air Canada/ American 1995 Canada-U.S. Bilateral 109.7%
1995
Montreal-Atlanta Delta 1995 1995 Canada-U.S. Bilateral 55.5%
Atlanta-San Jose CR Delta 1998 1997 U.S.-Costa Rica 118.5%
Dallas/Fort Worth- American 1996 Assignments of routes 336.6%
Santiago
Chicago-Hong Kong United 1996 (not daily) U.S.-Hong Kong Bilateral 21.1%
Chicago-London United 1995 U.S.-U.K Mini Deal, 1995 42.1%
Chicago-Sao Paulo United 1997 U.S.-Brazil, 1996 80.4%
Chicago-Buenos Aires United 1998 Reassignment of routes 41.1%
Houston-Sao Paulo Continental 1999 U.S.-Brazil, 1997 120.5%
Atlanta-Guadalajara Delta 1999 U.S.-Mexico, 1991 169.5%
Washington-Buenos United 2002 Reassignment of routes 208.7%
Aires
Washington-Sao Paulo United 2002 Reassignment of routes 88.4%
Detroit-Beijing Northwest 1996 U.S.-China,1995 174.3%
Dallas/Fort Worth-Lima American 1996 Assignment of routes 482.0%
Houston-Tokyo Continental 1998 1998 U.S.-Japan 116.6%
Atlanta-Rome Delta 1999 1998 U.S.-Italy 110.8%
Dallas/Fort Worth- American 2000 1995 Open Skies 115.3%
Zurich

Sources: United States Department of Transportation Databases 1B and 28IS; Statistics Canada Report 51-205, Air
Passenger Origin and Destination, Canada-United States Report, and InterVISTAS estimates.

Table III-2 understates the stimulation by using a strict year before/year after
timeframe. Traffic usually requires several years to adjust fully to a new service. Despite
the conservative approach, nonstop international services can often cause international
traffic to double in only a year, even for city-pairs that already have a profusion of one-
stop connecting services. Any mechanism that allows international services to proliferate
to non-traditional gateways can be a powerful stimulus to traffic. Restrictive bilateral
agreements, through confining service to a few named points, can thwart the growth
shown in Table III-2. They also exacerbate the airside and groundside congestion at the
largest gateways.

13
Figure III-2: Growth of United States-Canada Traffic, 1990-2004

20

18

16

14

12
Liberalized Market
10

1990 1992 1994 1996 1998 2000 2002 2004

Source: United States Department of Transportation Database 28DM

Figure III-2 depicts United States-Canada passenger flows. Until 1995, this market was
governed by a very restrictive bilateral, negotiated in 1966 and updated in 1974. It
forbade nonstop scheduled services on many important routes, including Toronto-
Washington, Atlanta-Montreal and Vancouver-Denver. Both countries had concluded a free
trade agreement in 1988, extended to Mexico in 1994. The February, 1995 air service
agreement allowed carriers of either nation to serve any route desired, although
temporary safeguards protected Canadian carriers. As shown by the graph, the previously
stagnant traffic saw rapid growth after liberalization.

C. AIR TRAFFIC AND ECONOMIC GROWTH

Tables III-1 and III-2 indicate a strong causal relationship between liberalization, air
service improvement, and international traffic. Table III-3 explores the final step of the
causal chain; the relationship between traffic and economic development. Many airports
have prepared economic impact statements to quantify their influence on their
communities. Several measures are used, including Gross Domestic Product, output,
employment, investment, and tax revenues. Several methods are available, and
assumptions vary widely between each such project. Despite the methodological
differences, the studies have reached a worldwide consensus that airports and civil
aviation can have an enormous and positive impact on regional prosperity. Table III-3

14
summarizes economic impact statements for a cross-section of airports and civil aviation
activities throughout the world.

Table III-3: Economic Impact of Civil Aviation

Passengers Employment Output Source


Des Moines, 1998 1.7 million 2,352 $182 million U.S. Des Moines International Airport
Newcastle, NSW .76 million 3,336 $540 million AU Newcastle Airport Limited
2005
Cincinnati, 2004 22 million 89,536 $5 billion U.S. University of Cincinnati
Reykjavik, 1998 1.8 million 1,156 11.4 Billion IKr University of Iceland Institute of
(2006) Economic Studies
Geneva, 1999 7 million 24,000 9.0 Billion SFr Aroports Internationale
Geneva
World Aviation, 2005 2 billion 29 million $2.96 trillion Air Transport Action Group
U.S.,
8% of world
GDP
U.K. Airports, 2004 229 million 580,000 22.2 billion Airport Operators Association,
gross value 2005
added
Toronto, 2001 28 million 138,000 $14 billion CD Greater Toronto Airports Auth.
Auckland, 2001 8.5 million 235,780 $14.2 billion NZ Auckland International Airport
All United States, 746 million 12.3 million $1.37 trillion U.S. Air Transport Association
2005
Inverness Airport, .5 million 2,297 120 million Inverness and Nairn Enterprise,
2005 Highlands and Islands
Enterprise and Highlands and
Islands Airports Limited
U.K. Air Freight 2,526,266 80,000- 4.99 billion U.K. Air Freight Study Report,
Industry, 2000 tonnes 100,000 U.K. Department of Transport

The instances shown in Table III-3 describe the impacts of facilities operating with a given
level of service. A further refinement involves measuring the incremental impact of a
specific improvement in air service. Table III-4 summarizes several studies that examined
specific instances of new flights or liberalization initiatives.

15
Table III-4: Incremental Impacts of Improved Air Service

Source Liberalized Market Impacts of Liberalization


Montreal, Toronto and Vancouver Airports Canada-U.S. Open Skies 455 jobs, $59.6 Cdn. Output, $31
million CD GDP
Impact of Liberalization on Hamburg, Liberalizing 10 country-pairs $1.3 million output, 126 jobs
1999.
Gillen and Hinsch
European Commission study by Brattle U.S-EU single market $8.1 billion in output, $5.2 billion
Group, 2002 benefits with lower fares
Denver International Airport Annual Impact of Daily Flight to $142 million impact
Asia
Metropolitan Washington Airports Impact of 4/week service to $325 million in activity
Authority Japan
Direct Exhibits of Federal Express, Impact of new FedEx U.S.-China 13,771 jobs, $1.6 billion
Docket OST 99-6323 Services expenditures
Baltimore Business Journal, November Mexican service, Baltimore- $54 million expenditures for
28 2005 Mexico City Maryland

Table III-4 again attests to the importance of new air service to economic development. In
most instances, such improvements are brought about merely by easing the constraints of
specific bilateral agreements, and allowing market forces to govern.

D. CATALYTIC EFFECTS FROM AIR SERVICE DEVELOPMENT

The relationships explored in Tables III-3 and III-4 view the various impacts as the
response of a pre-existing economy to incremental changes in civil aviation. They assume
that there will be no changes to the underlying structure of the regional economy. This
simplified approach is not appropriate for many situations. Growing evidence indicates that
new air services can lead to changes in the underlying structure of the regional economy,
through creating new capabilities, and forming a different set of transactor expectations.
These reactions can literally create new industries in a region, and allow the area to
compete for development throughout the world. These catalytic impacts are the most
difficult to quantify. Although most evidence is anecdotal, there is growing evidence that
these effects can be very large:

A 10 percent increase in the supply of intercontinental flights involves around a


4 percent increase in the number of headquarters of large firms located in the
corresponding urban area headquarters of knowledge-intensive sectors are

16
much more influenced by the supply of direct intercontinental flights than are
those of non-knowledge-intensive sectors.19

Nine foreign-owned companies in Northern Kentucky cited air service as an


important factor in their choice of location. The nine firms collectively employ
1,470 persons.20

In 2003, Kenya exported 50,000 tonnes of flowers by air freight.21

The growth of European air transport since 1995 has boosted European Union GDP
by 4 percent. The expected growth to 2025 will boost GDP of the 25 European
Union nations by a further 1.8 percent.22

Air service liberalization in Egypt could increase the GDP of the Travel and Tourism
industry by 12 percent by 2011, adding 260,000 full time jobs. Furthermore, the
total GDP for all sectors would increase by 1.8 percent.23

Between 1995 and 1999, the proportion of tourists who arrived in the Latin
American/Caribbean region by air jumped from 47percent to 55 percent.24

E. CONCLUSIONS

The evidence cited demonstrates a strong causal link between air service liberalization,
traffic growth and economic development. The next part of this section examines how
these interactions have influenced five specific markets. It lends further urgency to the
need for air service liberalization by illustrating the benefits of successful efforts.

Section IV of this study will summarize the development of an extensive model of


liberalization. It was designed utilizing a cross-section of hundreds of country-pairs
throughout the world. The resulting structure can be applied to any arbitrary country-pair.
The model provides further evidence of the benefits of a liberalized air service regime.

19
Germ Bel and Xavier Fegeda, Getting There Fast: Globalization, Intercontinental Flights and Location of
Headquarters, Unpublished, September 2005.
20
George Vredeveld, Marie Haney, Pooja Sharma and Anthony Apostolides, The Influence of International Airports
on Regional Economic Growth, Economics Research Group, Center for Economic Education, University of Cincinnati,
Cincinnati, 1997.
21
Oladele Samson Fatokun, African Air Transport in the 21st Century: A Case Study of the Contrasting Experience
of Nigeria and Kenya, MSc Thesis, Department of Engineering, Cranfield University, 2004.
22
The Economic Catalytic Effects of Air Transport in Europe, Eurocontrol Experimental Centre, Bretigny sur Orge
Cedex, 2005.
23
Aviation in Egypt: The Impact on Travel & Tourism, Jobs and the Economy, World Travel and Tourism Council,
2005.
24
The Contribution of Air transport to the Latin American/Caribbean Economy, Air Transport Consultancy,
International Air Transport Association, 2002.

17
F. AIR SERVICE LIBERALIZATION CASE STUDIES

1. GENERAL RESULTS
In order to confirm the premise that liberalization leads to market growth and economic
expansion, and to validate the results being generated by the economic model, we studied
five separate cases.

In each case, we examined the


US-UK Germany/UK-UAE
background of the bilateral
relationship, the history of traffic
growth, and its relationship to
benchmark parameters such as
GDP growth. In all cases
studied, it was apparent that,
depending on the size and
development of the economies,
there was substantial
Intra EU
incremental passenger traffic
and economic growth after air
Malaysia-Thailand
service agreements between the
countries had been liberalized.
In some cases, the liberalization
was of a transitional nature
that is from a rigid Bermuda I
type agreement to something
Trans-Tasman
less accommodating than Open
Skies, while in other cases the
liberalization was from a transitional to an Open Skies regime. In one case, we found that
liberalization occurred as a result of informal understandings between governments, with
no accompanying modifications to the formal air services agreement.

Post liberalization traffic growth tended to exceed pre-liberalization growth levels by


anywhere between 12 percent and 35 percent and up to 50 percent and greater,
depending on the periods measured. In all cases, the traffic growth produced significant
increases in economic output and job growth.

18
CASE STUDY

UNITED STATES UNITED KINGDOM MARKET

19
U.S.-U.K. AVIATION RELATIONSHIP A CASE STUDY

The United States and United Kingdom aviation relationship has, in many respects,
provided the origins of todays complex web of bilateral air services agreements (ASAs)
that frame trade in the field of commercial aviation. Beginning with the Chicago
Convention in 1944, when the United States tried and failed to sell its free market
principles to fifty world nations, there has been a tendency toward protectionism and
mercantilism led by the British. It was the British that vigorously opposed the U.S. free
trade position in 1944, and that led to a compromise two years later in Bermuda. Thus,
the Bermuda I standard ASAs that dictated how, when, and where air carriers could do
business, and at what price.

A. BACKGROUND

The aviation relationship between the United States and the United Kingdom has been
long, sometimes contentious. The U.S.-U.K. bilateral aviation agreement, while liberal in
many respects, retains strict limits on designations and operations between the U.S. and
the two principal U.K. airports, London Heathrow and Gatwick.

These throwbacks to protectionism have their roots in the 1920s and 1930s. However, in
August of 1944, the British government requested that the United States host an
international conference on civil aviation. The conference was convened in Chicago on
November 1, 1944, i.e. the Chicago Convention of 1944 when 54 nations met to establish
a post-war commercial aviation framework. There was much debate relative to the merits
of government control versus free trade principles. The U.S. supported what we essentially
refer to today as Open Skies. In that environment, carriers of any country would have
been free to operate on the basis of commercially driven decision-making relative to the
numbers of services they offered, the places they chose to fly to, and the prices they
would charge the traveling public.

The British supported an entirely different regulatory regime. It was the British position
that there should be an international regulatory body to administer a rigorous regulatory
framework. In part, the British rationalized their position on the basis that the U.S. would
be too powerful in the post-war economy. Moreover, they did not trust the marketplace to
make judgments, nor did they trust air carriers to make rationale commercial decisions. In
any event, the British preference was to regulate routes, carrier designations, market
entry and exit, frequency and size of services, and pricing. With these tools, governments
were thought to be able to control load factors, market shares, and profitability. At this
stage in the development of international aviation, the consumer was not a major
consideration.

The U.S., on the other hand, proposed to give the commercial entities discretion as to how
they would operate, and at what prices. In effect, the U.S. supported free trade in
aviation, although it was prepared to concede certain controls on capacity as a

20
compromise position. In addition, the U.S. (as it does today) supported a multilateral
regime, notwithstanding opposition from Pan American, an airline fearful of competition.

An interesting derivative of the Chicago Convention related to a recommendation of the


Canadian government that airlines establish rates and then file those rates with a regional
body. In addition, there was a recommendation that carriers should discuss the formation
of an international aviation organization. This, of course, led to the establishment of the
International Air Transport Association (IATA).25

The lack of agreement at the Chicago Convention led to the Bermuda Conference in 1946,
the first major step toward what was to become the traditional Bermuda I type bilateral
air service agreements.

That framework would be the basis for establishing grants of rights under ASAs over the
next 40 years. It also established IATA as the rate setting forum for international air fares,
but gave governments the right to reject fare offerings, should they be deemed
unacceptable, for whatever reason.

B. CAPACITY AND TRAFFIC ANALYSIS

In the years leading up to 1990, United States-United Kingdom traffic represented roughly
one-third of total U.S. to Europe traffic. Moreover, growth in the U.S.-U.K. market was
generally consistent with that of the U.S.-Europe market.

Capacity (flight frequencies and seats) and traffic grew in response to numerous factors
including growth in GDP, personal income, pricing, and liberalization initiatives with other
European nations.

When viewing the North Atlantic as a benchmark for the U.S.-U.K. market, we find that
the U.S.-Europe traffic exhibits an income elasticity between 1.0 and 1.5, depending on
the years measured.26 Growth in given country-pair markets also tended to vary as the
U.S. negotiated Open Skies agreements with individual European countries, beginning
with the Netherlands. Open Skies agreements tend to produce significantly enhanced
competition, more flights and seats, and lower fares. These factors combine to generate
more traffic to the market in question. Often that traffic is new to the market, i.e.
generative, due to the increased service levels and lower fares. However, those same
factors will often result in positive traffic diversion from other city or country-pair markets
where the regulatory regime is less favorable to the decision making flexibility that air
carriers need, and thus less favorable to the consumer.

25
Tanaja, Newal. U.S. International Aviation Policy, pg. 11.
26
European Liberalization and World Air Transport, Wheatcroft and Lipman, May 1990, pg. 127.

21
In a 1996 study for American Airlines, Economic Benefits of U.S.-U.K. Open Skies,
Roberts Roach & Associates states:

Wherever the industry remains regulated, each airline seeks to use the levers of
regulation to secure advantage for itself and to hobble its opponents. With the
U.S.-U.K. Open Skies, of course, one more regulatory forum will be closed;
everybody will be free to compete in the marketplace. Those who see an
opportunity to gain by manipulating the governmentand those who, by nature,
prefer regulationsee the alliance approval process as one more chance to seek
commercial advantage from the government rather than in the marketplace.27

During the mid-1990s, the U.S. and the U.K. agreed to liberalize the traditional Bermuda
II agreement that had been in place since 1978. As part of that agreement, the U.S.-U.K.
market was essentially deregulated, with the exception of Heathrow and Gatwick airports.
Otherwise, carriers could operate any city pairs, and at pricing that was commercially
determined. As a result, capacity and traffic grew considerably during the 1990s.

Capacity, as a result of the liberalization, grew by a compound annual rate of 7.8 percent,
and interestingly, resulted in a capacity (seat) share shift over the ten year period from 48
percent U.S. flag carriers in 1990 to 58 percent U.S. flag carrier in 2000. At the end of
2005, the U.S. still had a dominant share of the transatlantic market, with 56 percent,
notwithstanding the tendency of U.K. carriers British Airways and Virgin Atlantic to operate
more B747 and A340 aircraft. (Figure III-1)

27
Roberts, Roach & Associates, Economic Benefits of US-UK Open Skies, October 1996, pg. vii.

22
Figure III-1: Capacity Shares, U.S. vs. U.K. Carriers

Chart 1
Capacity Shares
US vs. UK Carriers

12,000

10,000
Number of Seats (000s)

8,000

6,000

4,000

2,000

0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005*

US Flag Capacity Total US-UK Capacity

Source: U.S. DOT T-100


*- Year-to-Date, October 2004 to September 2005

Traffic in the U.S.-U.K. market subsequent to the liberalization tended to grow at a rate in
excess of the general North Atlantic growth (Figure III-2). While that trend began to
reverse itself in the late 1990s, i.e. the North Atlantic grew at a faster rate beginning in
1999, that reversal may have been more a function of capacity limitations at Heathrow
and Gatwick, combined with the maturity of the U.S.-U.K. market subsequent to
liberalization.

23
Figure III-2: North Atlantic vs. U.S.-U.K. Traffic Growth

Chart 2
North Atlantic vs. US-UK
Traffic Growth

20.0%

15.0%

10.0%
% Change in Traffic

5.0%

0.0%
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005*

-5.0%

-10.0%

-15.0%

North Atlantic US - UK

Source: U.S. DOT T-100


*- Year-to-Date, October 2004 to September 2005

C. TIME LINE OF CAPACITY AND TRAFFIC

The early 1990s liberalization of the U.S.-U.K. market enhanced carrier choice, and clearly
led to healthy traffic growth. Exhibit III-3 provides a fifteen year history (1990 to 2005) of
traffic between the U.S. and U.K. and the U.S. and Germany. Interestingly, there is
roughly the same average annual growth in the U.S.-Germany market over the 15-year
period, 4.2 percent, as in the U.S.-U.K. market over the same time frame, 4.1 percent.

24
Figure III-3: U.S. to U.K. and Germany Traffic Growth

Chart 3
US to UK and Germany
Traffic Growth

20.0%

15.0%

10.0%
% Change in Traffic

5.0%

0.0%
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005*

-5.0%

-10.0%

-15.0%

US - Germany US - UK

Source: U.S. DOT T-100


*- Year-to-Date, October 2004 to September 2005

However, in the first 7-8 years subsequent to the liberalization of the U.S.-U.K. market,
the U.S.-U.K. market grew more rapidly than the U.S.-Germany market. It was only in
2000 and later that the German market began to grow more rapidly than the U.K. market.
This would appear to be attributable to other factors, such as capacity constraints at U.K.
airports.

D. ECONOMIC IMPACTS

The transitional liberalization that occurred in the U.S.-U.K. agreement in the mid-1990s
generated an incremental 9,197 full-time equivalent jobs in the United States and over
16,700 full-time equivalent positions in the United Kingdom. The Gross Domestic Product
of the United States expanded by $747 million; the United Kingdom saw a $970 million
increase.

The U.S. has been eager to negotiate an Open Skies agreement with the U.K., and that
could come about as a result of a U.S.-EU agreement.

25
In order to test the magnitude of such an agreement, we simulated the impacts of full
liberalization on the United States-United Kingdom market and found that it would result
in a 29 percent increase in traffic, some of which would derive from reduced fares, while
some would result from allowing any U.S. city to obtain nonstop service to Londons
Heathrow or Gatwick airports.

The economic benefits of this liberalization would be substantial. Over 117,000 full-time
equivalent positions would be created and the GDP of the two countries would grow by
roughly $7.8 billion.

The Roberts Roach study28, designed to support Open Skies as a vehicle to ensure
antitrust immunity for the American Airlines-British Airways alliance, generated estimates
of economic benefits to U.S. communities that could result from an Open Skies
agreement. In summary, the study concluded that Open Skies will open the floodgates to
dramatic increases in service by every one of the alliances existing competitors and by
any other U.S. or U.K. carrier that wants to enter the market. It will also permit access to
London Heathrow by any U.S. carrier.

The study further estimated the following over the first five-year transition period:29

An 86 percent increase in weekly flights between the U.S. and London

9.4 million new passengers between the U.S. and London

$9.1 billion in annual expenditures by foreign visitors in the U.S.

$3.6 billion in additional annual job-creating foreign investment in the United States

Thus, the potential results of further liberalization are substantial with


respects to enhancing the economies and job opportunities in both
countries.

28
Roberts, Roach & Associates, US-UK Open Skies, October 1996, pg. vii.
29
ibid pg viii.

26
CASE STUDY

INTRA-EUROPEAN UNION MARKET

27
INTRA-EUROPEAN COMMUNITY

A. BACKGROUND

This case study examines the impact of the legislation dating from 1992, which created
the so-called Single Aviation Market within the then twelve Member States of the
European Community. It covers:

(a) the impact of this deregulation on the Community market as a whole;

(b) the impact of this deregulation on a number of selected routes from Germany.

The process of legislation was spread over a period of six years. It was at the outset,
fiercely resisted, both by the industry and by many of the Member States. It is fair to say
that the adoption of the final package of legislation in 1992 created an entirely new
regime; a predominantly restrictive regime was replaced by a largely unregulated one. All
restrictions on cross-border routes within the EU were removed. Sweden, Finland and
Norway have had an agreement with the Community under which they adopted the
Communitys aviation legislation and were in effect part of the Community market.

The 1992 legislation was the third of three packages. The first two (1987 and 1990)
retained the bilateral system. In effect, by reaching agreement in the Council, the Member
States agreed that in their bilateral relations with each other they would observe rules
governing the following main areas:

1. The ability of governments to insist that capacity be tailored to avoid significant


departures from the 50-50 rule (capacity sharing or pooling). The first package
allowed a government to intervene if its airlines/airlines share of capacity fell
below 45 percent (and in a second step, 40 percent). The second package
liberalized the rules further, though again with limits beyond which Member States
could intervene.

2. A relaxation of fifth freedom restrictions, greater opportunities for multiple


designation and greater freedom to operate third/fourth freedom services, albeit
with qualifications.

3. Fares were still subject to control but the process shifted from a watered-down
system of double approval to double disapproval, with certain exceptions.

28
4. The freedom of airlines to enter into co-operative agreements was limited.
Revenue pools were restructured and the freedom of airlines to participate in the
IATA rate setting machinery was increasingly restricted.

The first two packages removed, or weakened, the provisions which enabled restrictionist
governments and airlines to thwart liberalization.

The changes permitted new fifth freedom routes from the U.K. to continental points. But
the time-scales involved, and the conservatism of many Member States limited the impact
of the changes. The period between them was short and discussions on the next steps
started very soon after each of the packages were adopted. The staging was not the result
of commercial pressures as much as a rapid development in the political acceptability of
an unregulated internal aviation market following in the slipstream of the creation of the
internal market for goods.

The 1992 package eliminated bilaterals for services within the Community. Its main
provisions were:

1. Community air carriers could exercise traffic rights on routes anywhere within the
Community. Until 1997 this included en route cabotage (e.g. Air France flying
Paris-Frankfurt-Berlin) provided that no more than 50 percent of the capacity was
used for the cabotage service. Freestanding cabotage was liberalized from 1997.

Safeguards protected routes where a public service obligation existed, particularly


thin routes operated by small aircraft. A provision enabled Member State A to
refuse access to a service using an airport in Member State B to which the airlines
of Member State A could not get access for reasons of congestion. It is also
possible to refuse or limit the use of traffic rights where serious congestion or
environmental problems exist. None of these safeguards has been extensively
used and they can be ignored for the purpose of this exercise.

Capacity limitations were made illegal except in cases of congestion or


environmental problems.

2. Community carriers were free to set air fares. Charter fares and seat and cargo
rates charged by Community air carriers were to be set by free agreement of the
parties to the contract of carriage.

3. Safeguards dealt with excessively low or excessively high fares. They have never
been applied.

4. Exemptions from the competition rules were provided for:

a) Joint planning and co-ordination of the schedule of an air service between


Community airports. This was to be subject to conditions: the results were to be
non-binding; it should be designed to ensure services at less busy times of day.

29
there should be no limitation of capacity; parties could not be prevented from
introducing additional services or withdrawing from the arrangement; it should not
seek to influence the schedules of non-participating carriers.

b) Joint operations on very thin routes and between two points where there had
been no direct service during the four traffic seasons preceding the beginning of
the joint operation; thinness was strictly defined.

c) Consultations on passenger and cargo tariffs. These were limited to fares to be


the subject of interlining arrangements; the consultations were not binding and did
not entail agreements on agents remuneration.

d) Slot allocation and airport scheduling.

To all intents and purposes, these changes came into force simultaneously,
throughout the Community, on 1 January 1993. The exceptions were that the
liberalization of freestanding cabotage became effective in 1997; the detailed
exemptions from the competition rules in June 1993. Their intent had been part of
the package agreed in 1992.

A tabular summary of the legislative developments follows on the next page.

Throughout the negotiating process there was considerable resistance from most of the
Member States and their airlines. The impact of the 1992 package can be illustrated by
considering the degree of liberalization that existed in the bilaterals which were in force
between the Member States immediately before the adoption of the package. In particular
the U.K. was the only country whose bilaterals systematically liberalized the rules on
capacity. When, therefore, the package came into force on January 1st, 1993, it
represented a greater change for the remaining Member States of the Community than for
the U.K.

B. CAPACITY AND TRAFFIC

Total Europe

The figures used for the EU market as a whole cover Geographical Europe30 rather than
the European Community. The Community traffic represents a large proportion of the
total. The data also cover only AEA members but they were usable until 1996 when the
low-cost operators began to develop. Thereafter we have applied to both ASKs and RPKs

30
As defined by ECAC, the route area Geographical Europe includes all scheduled services originating and
terminating within the 49 countries of the region comprising geographical Europe and European Russia up to the
Urals (longitude 55 E).

30
in order to account for the low-cost operators. This factor has been taken from a relevant
Commission study.31

Summary of Intra-EU Air Transport Packages32

1st package 2nd package 3rd package


Scope
From 1 January 1988 From 1 November 1990 From 1 January 1993
International scheduled passenger International scheduled passenger Intra-Community air transport
transport transport
Fare % of Fares Fare % of Fares approved by Free pricing
Fares types ref. approved by types ref. States However, provisions made for States
fare States Fully fare Unless double and/or the Commission to intervene
Discount Automatically flexible 106- disapproval against
Deep 66-90 Automatically Normal 95- Automatically -excessive basic fares (in relation to
Discount 45-65 Dbl approval econ 105 Automatically long term fully allocated costs)
All other Discount 80- Automatically -sustained downward development of
Deep 94 If dbl approval fares
Discount 30-
All other 79
Multiple designation by a State Multiple designation by a State allowed No longer applicable
Designation allowed if: if:
250,000 pass - 140,000 pass or 800 rt flight
st
(1 year after notification) (from Jan 91)
200,000 pass or 1,200 rt flights - 100,000 pass or 600 rt flight
nd
(2 year) (from Jan 92)
rd
180,000 pass or 1,000 rt flights (3
year)
Capacity shares between States Capacity shares of a State of up to 60% Unrestricted
Capacity
45/55% (from Jan 88) Capacity can be increased by 7.5% a
40/60% (from Oct 98) year
rd th rd th
3 /4 Freedom region to hub 3 /4 Freedom between all airports Full access to international and
Route routes permitted th domestic routes within the EU including
5 Freedom traffic allowed up to 50% of
Access th
5 Freedom traffic allowed up to capacity routes between states other than the
30% of capacity base of the carrier
Public service obligations and certain
th
Additional 5 Freedom rights for protection for new regional routes Exemptions for Greek Islands and
Irish and Portuguese rd th Azores
A 3 /4 Freedom service can be
Combination of points allowed matched by an airline from the other Cabotage is allowed for up to 50% of
Some exemptions State capacity if the domestic sector is
combined with a route to the home
Scope for traffic distribution rules and
country
restrictions related to congestion and
environmental protection Cabotage is unrestricted from April
1997
More developed public service
obligations and certain protection for
new thin regional routes
More developed scope for traffic
distribution rules and restrictions related
to congestion and environmental
protection

32
European Commission, 1996. The Single Market Review, Impact on Services, Air Transport. Brussels: Office for
Official Publications of the European Union.

31
For the purposes of comparison and to identify trends we have used three four-year
periods, i.e. 1990 to 1994, 1994 to 1998 and 1998 to 2002. We have only used the figures
of the first and last year of each four-year period. This enables us to ignore short-term
variations within the four-year periods such as, in the period 1990 to 1994, the impact of
the Gulf War (August 1990-March 1991). The figures for the third period are obviously
distorted by the effects of September 11, 2001. To achieve an undistorted trend figure, we
have taken the growth rate for the period 1998 to 2000 and applied it to the period 2000
to 2002. The adjusted figure for the third period for Geographical Europe is therefore
higher than the actual figure. However on one German route the adjusted level was lower
than the actual. The figures for the third period have obviously been affected by the
bankruptcy of Swissair and Sabena in 2001.

Table III-5 below and Figure III-7 illustrate the changes in ASK, passenger numbers and
operating RPK in the three four-year periods 1990 to 1994, 1994 to 1998 and 1998 to
2002.

Table III-5 Capacity and Traffic Growth for All Europe, 1990-2002

Increase in ASKs Increase in passengers carried Increase in operational RPKs

Market 1990- 1994- 1998- 1990- 1994- 1998- 1990- 1994- 1998-
33 34 35
1994 1998 2002 1994 1998 2002 1994 1998 2002

Total
Europe 27.5% 34.9% 36.2% 20.4% 37.8% 40.9% 23.6% 42% 35.2%

Source: Association of European Airlines

The legislation came into force on January 1, 1993. The 1990 package would itself have
caused some increase in the figures, although the change would likely have been modest.
We would not expect to see any real impact of the 1992 legislation before the summer of
1994 or possibly even 1995. We regard 1994 as a reasonable turning point on which to
base our analysis of trends. This led to the choice of the three four-year periods.

The most significant impact of the 1992 package was the stimulus to the low-cost airline
services. An operator could now order aircraft confident of a legally assured access to a
large market without legislative restrictions. This was (and is) a major encouragement to
investment and growth. Table III-6 below sets out the low-cost operators share of
capacity from 1996 to 2003. It will be clear from this that their impact on the market did

33
Corrected for September 11th, 2001.
34
Corrected for September 11th, 2001.
35
Corrected for September 11th, 2001.

32
not begin until 1996. Until around 2000, their presence was concentrated on routes from
the U.K.

Table III-6 Capacity Shares of Low Cost Carriers

Year Low-Cost Operators Share of


36
Capacity
1996 1.4%
1997 2.8%
1998 3.7%
1999 4.2%
2000 6.0%
2001 6.4%
2002 11.1%
2003 20.2%
Source: European Commission (2005). DG TREN: Analysis of the European Air Transport Industry 2003 (Contract
number B2-7040B-S07.17962). Brussels: Office for the Official Publications of the European Union.

Since July 2001 the full-service sector (including regional operators) has cut capacity in
the internal EU aviation market. July 2000 was the last point when full-service airlines
posted meaningful growth. The chart below37 (using actual figures rather than figures
corrected for September 11, 2001) demonstrates that from 2001 the growth of the low-
cost sector matched the decline of the full-service sector.

Figure III-6: Intro EU Capacity Growth


Intra EU Capacity Growth
Low Cost v Other Services, 1994 - 2003

1000000

800000
Change in Weekly Scheduled Capacity

600000

400000

200000

-200000

-400000

-600000
1995 v 1994 1996 v 1995 1997 v 1996 1998 v 1997 1999 v 1998 2000 v 1999 2001 v 2000 2002 v 2001 2003 v 2002

Low Cost Full service / Other

36
Percentage of weekly available seats provided by low-cost operators between 1996 and 2003; Source: European
Commission, 2005. DG TREN: Analysis of the European Air Transport Industry 2003 (Contract number B2-7040B-
S07.17962). Brussels: Office for the Official Publications of the European Union.

33
Once the impact of deregulation began to be felt, the operational RPKs and passengers
carried increased more than the ASKs. Traffic grew faster than capacity, indicating that
deregulation led to higher load factors and better aircraft utilisation. The low-cost
operators that entered the market after deregulation introduced lower fares which resulted
in higher load factors. Under pressure from the low-cost operators, the other carriers in
turn lowered their prices and increased their own load factors.

B. INDIVIDUAL ROUTES

Six specific routes within Europe demonstrate these general patterns. These routes
involved are from Germany on one hand to Italy, Belgium, France, Finland, Spain and the
U.K. Table III-7 below gives the growth in ASKs and passengers carried for the three
periods between 1990 and 2002.

Table III-7 Capacity and traffic Growth on selected Country-Pairs, 1990-2002

Rate of increase in ASKs Rate of increase in passengers carried

Country pair
38 39
1990-1994 1994-1998 1998-2002 1990-1994 1994-1998 1998-2002
10.3% 51% 45.2% 27.2% 49.1% 48.9%
Germany Italy
33% 47% 61.5% 39.6% 71.2% 37.6%
Germany
Belgium
25.5% 31.2% 23.7% 22.6% 44.3% 30.4%
Germany
France
23% 43% 16.2% 3.4% 47.7% 38.1%
Germany
Finland
61.7% 20.3% 25.5% 49.6% 28.3% 26.3%
Germany
Spain
16% 18.2% 44% 11.4% 18.7% 27.9%
Germany U.K.
Source: German Federal Statistics Office

The figures show that on the first four routes (i.e. Germany to Italy, Belgium, France and
Finland) the pattern matches that for Europe as a whole. The routes of course have
different characteristics. For Italy, Belgium and France, considerable traffic goes by car or
high-speed rail. Air travel is therefore more important for Finland than for the other three.

Two routes that did not follow this pattern are Germany-Spain and Germany-U.K. The
divergence for Spain reflects the large share of holiday traffic on charter flights. The

37
European Commission (2005). DG TREN: Analysis of the European Air Transport Industry 2003 (Contract number
B2-7040B-S07.17962). Brussels: Office for the Official Publications of the European Union.
38
Corrected for September 11th, 2001.
39
Corrected for September 11th, 2001.

34
figures for Germany-U.K. show that for the period 1990 to 1998, ASKs and passengers
carried grew at similar rates. The U.K.s bilaterals with Germany had already been
substantially liberalized before 1992. The new regime of January 1st, 1993, therefore made
much less difference to routes from the U.K. than to routes elsewhere in the Community.
The sharp upturn of capacity in the third period (1998-2002) in our view represents the
development of the low-cost carriers. Their share of total capacity in Europe as a whole
was 3.7 percent in 1998, but this was almost wholly concentrated on routes from the U.K.
hence the increase in the rate of growth of capacity during the period beginning in 1998.

Of the six routes from Germany, four saw rapid growth of both ASKs and passengers
carried in the period 1994 to 1998. This increase continued, though at a slightly lower
pace, in the period 1998 to 2002. The figures for total Europe show a similar pattern.

The contrast with the figures for Germany-U.K. and U.K.-EU15 is instructive. Until the
development of the low-cost airline sector in 1998 the growth of both ASKs and
passengers carried closely follows GDP growth and does not show the steep increase that
appears on the other routes and in the total figures for Europe.

The conclusion is that the sharp change in the regulatory regime of January 1, 1993,
resulted in dramatic increases in ASKs and passengers carried. This development started
after a time lag of four traffic seasons. The contrast between the growth for total Europe
and the German routes on one hand, and those for the U.K. on the other, suggests that
the liberalization was the predominant reason.

The other significant conclusion from these figures is that it took roughly five years for the
low-cost airlines to make a measurable impact. This impact was limited largely to U.K.
routes until 1998. It is only thereafter that the effect can be seen in the rest of the EU.
This has been the most dramatic consequence of the 1992 legislation; this is
demonstrated by the five-fold increase in the low-cost operators share of capacity
between 1998 and 2003. It may well be that the one factor that brought this about was
the contrast between the predominantly restrictive bilaterals before 1992 and the
predominantly liberal regime introduced simultaneously on every cross-frontier route in
the Community. The growth seen in the market as a whole for the first five years can not
be regarded as permanent. The subsequent growth in the capacity of the low-cost sector
was to be matched by a decline in the capacity of the full-service sector.

C. TREND IN CAPACITY AND TRAFFIC

The essential events driving changes in traffic and capacity trends are:

1. 1993: Coming into force of the legislation creating a liberalized internal aviation
market in the EU (routes between Member States);

2. 1996: First impact of low-cost carriers mainly on routes to and from the U.K.

35
3. 1997: Extension of liberalization to routes within each Member State.

Figure III-7, Rates of Growth of ASK and Passengers for Europe as a Whole, Compared to
the EU 15s GDP, 1990-2002

Graph 1 Rates of growth of ASK and PAX for Europe as a whole, compared to the EU 15's GDP, 1990
- 2002

350 350

300 300

250 250
Index = 100 in 1990

200 200

150 150

100 100

50 50

0 0
1990 1994 1998 2002

PAX Europe GDP EU 15 at current market prices (2005) ASK Europe

Sources: European Commission, World Bank

As indicated earlier in this paper, the majority of the European flag carriers were initially
skeptical of the consequences of liberalization. Only BA and KLM supported it, though
eventually SAS joined them. Once the legislation had been adopted, all airlines had to
adjust. For some the abolition of the protection they had hitherto enjoyed resulted in a
series of applications for state aid. In response the European Commission agreed to these
applications on the basis of first time last time i.e. aid was allowed to facilitate
adjustment to the new situation but thereafter was banned.

By and large the flag carriers have used the abolition of controls on capacity and fares
more than they have the freedom of access to routes. The creation of a single market also
facilitated the development of alliances among the European airlines (or their membership
in the global alliances that came into being at that time) and later of outright mergers
(e.g. Air France/KLM).

36
The most marked development, however, has been the growth of the low-cost carriers.
They started in the U.K. and are now spreading throughout the 25 Member States. This
has transformed the internal market and forced changes of approach by the flag carriers.
The impact of these changes on the European airline industry is far from complete.

D. ECONOMIC IMPACTS

The liberalizations that created the Single European Market dramatically increased intra-
European air travel. The 44 million additional passengers attributable to the new market
regime represented an increase of fully 33 percent. The additional traffic could only be
accommodated by 681 new daily return flights.

The traffic expansion spurred development of both the travel/tourism sector and other
industries. Fully 1.4 million full-time equivalent jobs resulted from the liberalization, and
the European GDP grew by $US 85 billion.

37
CASE STUDY

UNITED ARAB EMIRATES-UNITED KINGDOM AND GERMANY MARKETS

38
UAE TO THE U.K. AND GERMANY

A. BACKGROUND

We have examined the two routes together, although the timing of the introduction of
liberal regimes differed. Liberalization of the Germany-UAE regime took place in one step
in 1986. Multiple designation was allowed, limits on frequencies and capacity were lifted;
choice of aircraft type and schedule were freed; between and beyond points and fifth
freedoms were unlimited. The only limitation was on the number of German points that
could be served by UAE carriers.

The U.K.s liberalization of 1998 replaced a restrictive bilateral. The regime negotiated in
1998 was the result of a policy decision taken by the U.K. in the early 1990s to liberalize
its bilateral arrangements unless there were good reasons not to do so. The regime from
1998 removed all restrictions on flight frequencies, capacity and fares. Fifth freedom rights
were available to other Gulf points and Saudi Arabia; but not between the U.K. and North
America (where competition was substantially more aggressive) nor from the UAE
eastwards.

B. CAPACITY AND TRAFFIC

To avoid the distortion caused by the effects of September 11th, 2001, we have taken the
growth rate for the period 1998-2000 and applied it to the period 2000-2002. The figures
for the third period are thus higher than the actuals.

Liberalization of the Germany-UAE market in 1986 resulted in high growth rates (both in
terms of ASKs and passengers carried), although the absolute traffic figures are not large.
The smaller absolute figures suggest treating the growth rates before 1990 with caution.

Although the U.K.-UAE route was liberalized much later, the commercial situation on the
two routes in 1998 was similar. There were only nine weekly flights by Emirates / Gulf Air
between Germany and the UAE. From the United Kingdom, British Airways ran a twice
daily service, as did Emirates and Gulf Air. In 1998, Emirates was beginning its expansion,
and Virgin Atlantic was pressing for Gulf route authority.

Table III-8 below displays similarities for both markets. Figures III-8 and III-9 summarize
the increase of ASK and passengers carried on U.K.-UAE and Germany-UAE respectively,
against GDP of the EU15 for the period 1990-2002. Graphs 3 set the rates of increase of
ASK and passengers carried on U.K.-UAE routes, compared to GDP of the EU15 and GDP
of the UAE40 for the period 1994-2002.

40
Based on Staff Country Reports of the International Monetary Fund (IMF).

39
The larger growth for Germany for the interval to 1998 reflects the earlier date of
liberalization. The statistics for Germany-UAE start from a lower base than the figures for
U.K.-UAE and, despite the steeper rise in the Germany-UAE graph from 1998 onwards, the
Germany-UAE market is still smaller than the U.K.-UAE market.

The largest change in both cases occurred after 1998. This is attributable to the growth of
Emirates Air. Their fast expansion (though not of course their style of operation) is similar
to the low-cost carriers in the intra-EU market. It supports the view that the liberalizing of
the bilaterals was a necessary condition for the expansion in activity.

Table III-8 Capacity and Traffic Growth, United Kingdom/Germany-United Arab Emirates,
1990-2002

Rate of increase in ASKs Rate of increase in passengers carried


Country pair 41 42
1990-1994 1994-1998 1998-2002 1990-1994 1994-1998 1998-2002

Germany UAE 22.7% 2.3% 77.1% 118% 59.8% 117%

U.K. UAE 29.8% 37% 129% 26% 52% 125%

Sources: German Federal Statistics Office, United Kingdom Civil Aviation Authority

This analysis shows that the major impact of liberalizing the two bilaterals results from the
expansion of Emirates and the development of Dubai as a hub for traffic between Europe
on one hand and the Far East and Australasia on the other. The rise of Dubai as a
financial, trading and tourist center alone would have supported more modest growth. It is
reasonable to conclude that the liberal bilateral arrangements between the UAE and the
two major financial centers of the EU contributed to the confidence which is necessary for
such developments.

C. TIME LINE PLOT OF CAPACITY AND TRAFFIC

We have ignored figures before 1990, since although the Germany-UAE bilateral
arrangements were liberalized in 1986, the absolute figures were so low that the figures of
percentage growth would be misleading.

41
Corrected for September 11th, 2001.
42
Corrected for September 11th, 2001.

40
Figure III-8

Graph 1 Rates of increase of ASK and PAX for the UK - UAE route compared to the EU 15's GDP,
1990 - 2002

800 800

700 700

600 600

500 500
Index= 100 in 1990

ASK UK- UAE


400 400 GDP EU 15 at current market prices
PAX UK - UAE

300 300

200 200

100 100

0 0
1990 1994 1998 2002

Figure III-9:

Graph 2 Rates of increase of ASK and PAX for the Germany - UAE route compared to the EU 15's
GDP, 1990 - 2002

800 800

700 700

600 600
Index = 100 in 1990

500 500

ASK Germany - UAE


400 400 GDP EU 15 at current market prices
PAX Germany - UAE

300 300

200 200

100 100

0 0
1990 1994 1998 2002

41
Figure III-10:
Graph 3 Rates of increase of ASK and PAX for the UK - UAE route compared to the EU 15's and the
UAE's GDP, 1994 - 2002

350 350

300 300

250 250
Index = 100 in 1994

200 200 ASK UK - UAE


GDP EU 15 at current market prices (2005)
GDP UAE at current market prices
150 150 PAX UK - UAE

100 100

50 50

0 0
1994 1998 2002

D. ECONOMIC IMPACTS

Germany-UAE

The 1998 liberalization, combined with a rising price of oil and rapid expansion by
Emirates Air and later, Etihad, prompted a 59 percent increase in total United Kingdom-
United Arab Emirates traffic. By 2005, total country-pair traffic exceeded 3.2 million
passengers, of which 1.1 million could be attributed to the more liberal air service
agreement.

Both nations benefited significantly. In the United Arab Emirates, over 5,300 full-time
equivalent positions were created, and GDP expanded by over $110 million. The expansion
created over 18,700 full-time equivalent positions in the United Kingdom, and over one
billion dollars additional GDP.

Germany-United Arab Emirates

In 2004, over 1 million persons traveled between Germany and the United Arab Emirates.
The liberalization of 1986, through allowing a proliferation of direct services between the
two countries, prompted a market expansion of almost 167,000 passengers. Traffic was
fully 19.7 percent higher than it would have been in the absence of the liberalization. The

42
additional traffic required a further two flights daily in each direction, using 150-seat
aircraft and a 75 percent load factor.

The expanded traffic benefited both nations. In the United Arab Emirates, the equivalent
of 745 new full-time positions was created, with a $15 million increase in GDP. The 2,600
new full-time equivalent positions in Germany accompanied a $152 million growth in GDP.

43
CASE STUDY

AUSTRALIA-NEW ZEALAND (TRANS-TASMAN) MARKET

44
A. BACKGROUND

This case study traces the liberalization of aviation in the Trans-Tasman market between
Australia and New Zealand. These two countries have been closely linked through British
Commonwealth colonial ties and, due to their comparative isolation geographically; have
consistently worked to liberalize government aviation policies.

The first steps towards economic liberalization can be traced to 1966 when the New
Zealand and Australia Free Trade Agreement was signed. This agreement was in place for
17 years until March 28, 1983 when the Australia-New Zealand Closer Economic Relations
Free Trade Agreement (ANZCERTA) was concluded. The ANZCERTA set a foundation as an
innovative agreement, which not only created a liberal business and economic regime for
goods and services, but also set a collaborative umbrella to deal with customs, transport,
regulatory, product standards and business law issues. The ANZCERTA established a
market that maintains one of the most open economic trade relationships between any
two countries in the world. The ANZCERTA is continually reviewed to ensure that the
agreement remains effective in all sectors of the economy.

This analysis will focus specifically on the effects of the Single Aviation Market that was
established in 1996, preceding full Open Skies in 2000.

Single Aviation Market

Australia and New Zealand concluded a Single Aviation Market (SAM) agreement, effective
as at November 1, 1996. The goal of the Single Aviation Market was to bring the two
countries closer together within the elements of the ANZCERTA. The main components of
the agreement included the opening of ownership and control regulations in the bilateral
market, the introduction of unlimited frequencies for Trans-Tasman services and a
provision which allowed airlines of either country to operate domestic flights within the
other country. While the SAM agreement opened up many new opportunities within the
Trans-Tasman market, it did not address beyond markets to third countries. Those
markets were still under the original 1961 Australia New Zealand Air Services Agreement
and the subsequent 1992 Memorandum of Understanding. Two different labels of air
carriers were created from the agreement: the Domestic and the SAM airline. The
Domestic airline designation allowed carriers to fly domestic services in each others
domestic market and the SAM designation harmonized ownership, control, technical and
safety certifications from each countries regulatory agencies.

During this time, Air New Zealand held an equity position in Ansett Australia and British
Airways a position in Qantas. The Single Aviation Market agreement broke barriers in the
carriage of cabotage traffic, created ownership and control flexibilities, and deregulated
capacity, designations, and frequencies. More importantly, the SAM agreement laid the

45
foundation for a more liberal agreement that, in the future, would open markets beyond
the Trans-Tasman.

Open Skies

The Australia New Zealand Open Skies agreement was initialed in November of 2000
and entered into force on August 8, 2002. This agreement removed the last substantive
restrictions within the bilateral air services market and served as the culmination of a truly
open bilateral air service market. There were no longer any restrictions on flights to,
within, and beyond the territory of the other party. Before the Open Skies agreement,
beyond rights were limited by weekly frequencies. Seventh freedom cargo operations were
included to help extend the single market outwards in the air cargo market. New beyond
markets brought greater capacity on the Trans-Tasman as new international connections
were created between major cities.

B. CAPACITY AND TRAFFIC

Through liberalization, many new routes have been started by both network flag carriers
and low coast carriers on both sides of the Trans-Tasman. In additional to new routes, the
liberal market environment has allowed for new secondary gateways to compete for
traffic, especially in New Zealand. Figures III-11 and III-12 compare nonstop routes
between 1995 and 2006 and illustrate the fragmentation that has taken place in this major
market for both business and leisure traffic.

Since 1994, capacity and traffic on services across the Tasman Sea has shown steady
growth by most air carriers. While liberalization has facilitated growth, there have not
been sharp increases in services. The liberalized agreements have established a regime in
which the Trans-Tasman market became a breeding ground for competitive air service by
carriers in the bilateral market, and also from third country carriers from the Middle East,
Southeast Asia and North America. Carriers such as Freedom Air and Pacific Blue have
opened new Trans-Tasman routes to alternate secondary airports and tourist destinations.
The flag carriers and incumbents, Air New Zealand and Qantas, have remained dominant
forces in terms of passenger growth and market share.

Liberalization in the Trans-Tasman example has resulted in fragmentation of the market


and new opportunities for low cost carrier services. With fewer restrictions on frequencies,
ownership, control, designations and beyond rights, carriers have been able to expand.
This environment has permitted flag carriers to connect the Trans-Tasman to their global
networks, while giving low cost carriers flexibility to add markets and frequencies for
market presence.

The full spectrum of high volume, tourist-oriented and secondary destinations have grown.
Figure III-13 illustrates the ten year trend. The compound annual growth rate (CAGR) in

46
passengers in non-major markets exceeded the growth in major markets, which is
frequently the result when liberal agreements allow added competition and permit entry
by low fare carriers. The growth of secondary city-pairs was largely driven by Freedom Air
and Pacific Blue.

Figure III-11: Trans-Tasman Nonstop Services - April 1995

Cairns

Brisbane

Perth
Sydney
Melbourne
Auckland
Hobart

Wellington

Christchurch

Trans-Tasman
Nonstop Services
April 1995

Source: Official Airline Guide

47
Figure III-12: Trans-Tasman Nonstop Services - April 2006

Cairns

Brisbane
Gold Coast

Perth
Adelaide
Sydney
Melbourne
Auckland
Hamilton
Palmerston
Wellington
Christchurch
Dunedin
Queenstown
Trans-Tasman
Nonstop Services
April 2006

Source: Official Airline Guide, Freedom Air website

48
Figure III-13:

Analysis of Growth in Major Markets vs. Non Major Markets


1994 - 2004
Australian Bureau of Transport and Regional Economics
4,500,000 600,000

4,000,000
500,000
3,500,000
CAGR for Major

Total Passengers - Non Major Markets


Total Passengers - Major Markets

Markets 7.10%
3,000,000 400,000

2,500,000
300,000
2,000,000
CAGR for Non
Major Markets -
1,500,000 14.45% 200,000

1,000,000
100,000
500,000

0 -
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Major Markets Non Major Markets

Source: Australian Bureau of Transport and Regional Economics, International City Pair Data
Major Markets includes all routings between main New Zealand markets (Auckland, Christchurch, Wellington) and
Australian markets (Sydney, Melbourne, Brisbane, Adelaide, Perth)

C. MARKET TRENDS

The 1996 SAM agreement and the 2000 Open Skies agreement between Australia and
New Zealand provided the Trans-Tasman with a liberal government regime for air
services. Figure III-14 illustrates total passenger by uplift/discharge in the Trans-Tasman
market and Figure III-15 analyzes top countries served by passengers from 1999-2005.

49
Figure III-14:

Total Passengers by Uplift/Discharge


Trans-Tasman Market
Australian Bureau of Transport and Regional Economics
YE Aug 1995 - YE Aug 2005

2,000,000

1,800,000

1,600,000

1,400,000
Total Passengers

1,200,000

1,000,000

800,000

600,000

400,000

200,000

0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Air New Zealand Emirates Freedom Air International Pacific Blue Qantas Airways United Airlines

Source: Australian Bureau of Transport and Regional Economics, Airline Data Tables
Data is annually year ended August 1995 year ended August 2005

The post Trans-Tasman liberalization data demonstrates that different types of carriers
have been able thrive in a deregulated environment. Dominant flag carriers have indeed
survived the two liberalizing agreements by growing their traffic and maintaining their
market position. Liberal beyond market rights have also helped connect Trans-Tasman to
international networks of both Qantas and Air New Zealand.

All of the current carriers in the Trans-Tasman market have experienced growth over
many years. Liberalization has clearly changed the dynamics of the Trans-Tasman. Not
only have Air New Zealand and Qantas had to continually adapt to new low cost and third
country carriers in the market, but the consumer has seen new opportunities for travel,
both leisure and business. This is made possible by the liberal aviation policies that govern
the Trans-Tasman market.

50
Figure III-15 compares total passenger numbers to international destination. This chart
highlights the importance of the Trans-Tasman routes to both countries. New Zealand has
always remained the top destination for air travelers leaving Australia. Singapore, another
country known for liberal aviation policies, also shows growth in Chart 5. Other countries
such as Japan, Hong Kong, United States and U.K. all illustrate constant service to the
Australian market with increases and decreases due to changes in market air carriers.

Figure III-15:

International Passengers by Uplift/Discharge Country


1999-2005
Australian Bureau of Transport and Regional Economics

6,000,000

5,000,000

4,000,000

3,000,000

2,000,000

1,000,000

-
1999 2000 2001 2002 2003 2004 2005

New Zealand Singapore Japan Hong Kong USA UAE UK

D. ECONOMIC IMPACT

By 2005, Trans Tasman traffic was fully 56 percent higher than it would have been in the
absence of any liberalization. The relaxed market controls had increased total traffic by
over 1.7 million passengers per year. The additional volume would require a further 27
flights daily in each direction, using 120-seat aircraft.

The economic impact methodology uses similar coefficients for both Australia and New
Zealand. Each nation gained more than 20,600 full-time equivalent positions from the
liberalization and the ensuing traffic increase. The GDP of each country grew by $726
million U.S.

51
CASE STUDY

MALAYSIA THAILAND

52
A. BACKGROUND

The following case study deals with overall liberalization in the Malaysia-Thailand market.
The two Southeast Asia countries sit at a crossroads of global travel. While there is little
data available for a thorough analysis of the market, by examining rights within the
original ASA and tracking overall capacity trends, one can draw conclusions about the
impact of liberalization on the Malaysia-Thailand market.

The original agreement between the governments of Malaysia and Thailand was signed in
1969. This agreement type (Bermuda I) is still seen in many bilateral relationships around
the world today. It allows named points, carrier designations, frequencies and capacity to
be added in the market based on government consultations. While more restrictive than
the American style Open Skies, the agreements Bermuda I provisions allow new services
to be introduced to the market. Subsequent Memoranda of Understanding (MOUs)
between Malaysia and Thailand have allowed services and frequencies to increase, thus
driving an increase in overall traffic.

Many attributes of a liberalized market can be found within the terms of the agreement
between the countries. They include no restrictions on points served, multiple
designations, code sharing rights and open frequencies. Fifth freedom, intermediate or
beyond, and seventh freedom cargo operations are included in the agreement. As in most
other bilateral agreements, cabotage is not included.

Malaysia and Thailand have numerous air service agreements with third countries. They
have both concluded Open Skies agreements with the United States. Both currently
support a multiple designation policy as a way of increasing their countries role in global
tourism and trade. Liberal aviation policies will help both nations develop huge
investments in new airports for Bangkok and Kuala Lumpur. With regards to points served,
Malaysia and Thailand have open policies. The key restriction in the Malaysia-Thailand
bilateral agreement covers beyond rights and seventh freedom cargo operations. Code
sharing rights have allowed Malaysia and Thai Airways to cooperate routes between the
two countries. In addition, both countries promote secondary international destinations for
tourism (Thailand: Phuket, Chiang Mai; Malaysia: Langkawi).

53
Role of ASEAN

The Association of Southeast Asian Nations (ASEAN) was formed in 1967 by Thailand,
Indonesia, Malaysia, Singapore and the Philippines. ASEAN serves as a regional bloc,
similar to the European Union or the North American Free Trade Agreement (NAFTA). It
works to harmonize policy and encourages cooperation on trade, tourism, and economic
growth. ASEAN now includes Brunei, Vietnam, Laos, Myanmar and Cambodia. ASEAN has
established an end-goal of full economic integration by 2020.

Within ASEAN, other sub groups have been created such as the BIMP-EAGA43 group and
the IMT-GT groups44 to further enhance regional cooperation on trade and tourism. One of
the trade areas closely analyzed by the ASEAN countries is air transport. In November
2004, all 10 ASEAN member countries signed the ASEAN Framework Agreement for
Integration of Priority Schedules. This agreement will allow for a phased approach to
ASEAN Open Skies. The plan includes unlimited point-to-point operations between ASEAN
capital cities in 2008 and unlimited fifth freedom operations from those same cities in
2010. Full cargo open skies are planned for December 2008. The regional Open Skies
initiative will also have synergies with other bilateral Open Skies agreements.

To understand fully the Malaysia-Thailand market it is important to examine the key


factors that drive governments toward liberalization, including airport development,
national flag carriers, low cost carriers and inbound tourism.

B. CAPACITY AND TRAFFIC

Malaysia-Thailand capacity and traffic have changed only when new carrier entry changed
the competitive landscape. For many years, the Malaysia-Thailand market was dominated
by Malaysia Airlines and Thai Airways. Due to flexibilities in the air service agreement,
both were able to serve any points in each others country while maintaining capacity at
levels that were profitable for both flag carriers. Malaysia and Thailand separately had
agreements with third countries that accommodated carriers on the Kuala Lumpur-
Bangkok route as fifth freedom operators. In the last 10 years, carriers such as Lufthansa,
Qatar Airways, and Gulf Air have served this route on a fifth freedom basis.

The ASA allows the flag carriers to enter into cooperative marketing agreements (i.e. code
sharing). The two carriers have had code sharing operations for many years, creating
control over capacity and traffic growth.

43
Brunei Darussalam-Indonesia-Malaysia-the Philippines East ASEAN Growth Area.
44
Indonesia-Malaysia-Thailand Growth Triangle.

54
In 2004, AirAsia entered the Malaysia-Thailand market causing a surge in capacity (seats)
in the market. AirAsia was targeting the third and fourth freedom markets, i.e., traffic
between the two countries.

The bilateral agreement between Malaysia and Thailand permitted AirAsia to expand in the
market by new designations and frequencies.

Figure III-16:

Total Monthly Roundtrip Seat Capacity


Malaysia - Thailand Nonstop Scheduled Services

140

120

100
Total Monthly Roundtrip Seats

80
(000s)

60

AirAsia begins
40
Service between
Malaysia and
Thailand
20

0
Feb- May- Aug- Nov- Feb- May- Aug- Nov- Feb- May- Aug- Nov- Feb- May- Aug- Nov- Feb- May- Aug- Nov- Feb- May-
01 01 01 01 02 02 02 02 03 03 03 03 04 04 04 04 05 05 05 05 06 06

Source: OAG Schedules February 2001 May 2006

The above chart summarizes the increase in capacity in the Malaysia-Thailand market,
driven by AirAsia. AirAsias low cost structure helped spur traffic while generating
increased competition from the incumbent flag carriers.

The growth noted in Figure III-16 is a partial result of the agreement between Malaysia
and Thailand. The agreement is not restrictive in terms of allowing new entry into the
market. The agreement does maintain the governments role in responding to market
demand. AirAsias interest in serving international markets prompted the government of
Malaysia to open designations, frequencies and points in its bilateral agreements. This

55
regime differs from the U.S. Open Skies model in that it does not remove the government
from commercial marketplace decisions.

C. TRENDS IN CAPACITY AND TRAFFIC

Figure III-17: Malaysia-Thailand Nonstop Services 1996

Chiang Mai

Malaysia - Thailand
Nonstop Services 1996

Bangkok

Phuket
Hat Yai

Penang

Kuala Lumpur (Subang)

Source: OAG Schedules

Figure III-17 shows service in the Malaysia-Thailand market in April 1996. Most of the
routes and capacity remained concentrated around Bangkok and Kuala Lumpur. Routes to
tourist destinations were being served on both sides of the border. Service was largely
dominated by Malaysia Airlines and Thai Airways.

56
Figure III-18: Malaysia Thailand Nonstop Services 2006

Chiang Mai

Malaysia - Thailand
Nonstop Services - 2006

Bangkok

Koh Samui

Phuket

Hat Yai

Kota Kinabalu
Penang

Kuala Lumpur (Subang)


Kuala Lumpur (KLIA)

Source: OAG Schedules

When analyzing the updated map for April 2006, conclusions can be drawn about the
liberalization effects. The total number of routes in Figure III-18 reflects new low cost
carrier service in the market and the construction of the Kuala Lumpur International
Airport (KLIA). As AirAsia grew domestically it also develop an international network to
destinations such as Kuala Lumpur-Bangkok/Phuket and Bangkok-Penang. While these
routes were already served in 1996, the total capacity has grown due to AirAsias low fare
presence. Fragmentation has also occurred as evidenced by new routes such as Bangkok-
Kota Kinabalu. A passenger seeking to fly this route on a flag carrier would have had to
travel over a connecting hub, but as shown in Figure III-18, new point-to-point services
offered by AirAsia have developed secondary markets.

As the Southeast Asian market has grown, Malaysia Airlines and Thai Airways have had to
change their overall strategies. AirAsia continues to every a major influence on the flag
carriers. However, third country carriers have had little effect in the Malaysia-Thailand
market.

57
D. ECONOMIC IMPACT

In 2005, 1.3 million passengers traveled between Thailand and Malaysia. Of this total,
over 370,000 can be attributed to the combination of the liberalized regime and the entry
of a new low cost carrier. This suggests that the direct and indirect effects of liberalization
have caused a market expansion of over 37 percent. To carry the additional traffic,
airlines would have needed to operate an additional 5 daily flights in each direction.

The economic and tourism impacts of Thailand and Malaysia are identical. Each nation
obtained more than 4,300 full-time equivalent positions and a stimulus of over $114
million to their respective GDPs.

58
The Economic Impact of
Air Service Liberalization

IV. The ModelDescription


& Methodology
IV. THE MODELDESCRIPTION AND METHODOLOGY

A. INTRODUCTION

A mathematical model of air service liberalization must accommodate a wide range of


regulatory change, involving nation-pairs and airlines with greatly varying economic and
demographic conditions. With over two hundred sovereign nations, it must apply to over
40,000 country-pairs. It must distinguish between total and partial efforts of liberalization,
and must express the socioeconomic consequences using a method, and a series of
parameters, with worldwide relevance.

This chapter summarizes the methods for estimating the consequences of air service
liberalization on passenger traffic, air freight traffic, employment, gross domestic product,
tourism and catalytic impacts for any country-pair. It tests the hypothesis that restrictive
air service agreements constrain air commerce, and impose undesirable consequences on
the economies of the signatory nations. This chapter summarizes the two methods
employed in this study, and how they together provide statistically valid and economically
relevant conclusions on air service liberalization.

B. OVERALL METHODOLOGY

Alternate Approaches for Evaluating the Impact of Liberalization on Traffic

Two methods can quantify the benefits of liberal bilateral agreements. Both depend on
comparing liberalized regimes to more restrictive ones. They use widely different
mechanisms for holding other variables constant. Both are analytically useful, but present
a wide range of research challenges. The two methods together provide the means to
trace the complicated and situation-specific dynamics of the process, yet generalize it to
the most restrictive aviation regimes.

1. TIME SERIES OR CASE HISTORY


This method calls for selecting one or more representative country-pairs which have
undergone recent liberalization efforts. The impact of the change on the level of traffic can
be measured by comparing the levels before and after. Using the same country-pair
helps satisfy any other-things-being-equal requirements, and isolates the factors specific
to liberalization.

The five case histories represent an application of this method. However, they also
highlight the pitfalls of generalizing any such findings to the worldwide level. The major
issues to extending a case history analysis to a generalized and widely applicable model
include:

59
No two cases are identical. While there is general agreement on what constitutes a
liberalized market,45 the initial, pre-liberalization conditions can be radically different in
countless ways, most of which are impossible to measure.46 It is therefore very difficult
to relate a particular quantity of traffic growth to a particular level of liberalization. The
experience of one isolated and unique case cannot necessarily be generalized to others.

Even if two nations do liberalize an agreement, both parties will not necessarily make
full use of its provisions. Many will not assign a second airline, even if fit, willing, and
able, to an international route even if authorized to do so by the agreement

Some nation-pairs may have highly restrictive bilaterals, creating a very uncompetitive
market de jure. However, they may, on a discretionary basis, allow numerous
deviations, leading to a de facto liberalization.

Sometimes, the two nations are too small, too far apart, or affected by other
differences that preclude profitable international air services even under the most
favorable bilateral agreements. No traffic growth would result, at least in the
immediate term.

Fifth freedom rights. A country-pair may be served, not only by the airlines of the two
respective nations, but by airlines of third countries. They are granted revenue traffic
rights only if the bilateral agreements between the third nation and both members of
the country-pair include the necessary provisions, although such services face many
formal and informal restrictions.47

The airline industry is notoriously volatile. Traffic and earnings are very sensitive to
economic cycles, and to geopolitical events such as the 9-11 terrorist attacks on
Washington and New York, and the Severe Acute Respiratory Syndrome (SARS) of
2003. Structural changes in the industry, such as airline mergers, failures and strategic
alliances create further complications.

Traffic will often grow despite artificial restrictions, because of economic growth.

45
The classical open skies definition of a bilateral agreement would allow any airline from either signatory nation
to serve any city-pair in any combination. Each airline would be free to set its fares, choose the capacity it wishes to
provide, and carry revenue traffic to any third country permitted by the applicable bilateral agreements.
46
The initial circumstances could reflect a vast combination of feasible route definitions, capacity controls, pricing
distortions, operating time constraints, limits affecting on-airport flight handling, currency repatriation and other
concerns.
47
Some bilateral agreements, if they do not prohibit such operations altogether, specifically limit fifth freedom
capacities or traffic volumes. The airline exercising the rights is usually forbidden from undercutting the designated
airlines of the two nations. Although a hard right, obtained often through intense negotiation, such provisions
considered to be partially a privilege. There is often an informal and unspoken understanding that the airlines
exercising these rights do so only on a fill-up basis, clearly subordinated to services involving their home countries.
They are not expected to inflict commercial damage on the airlines of the country-pair. There are many instances of
fifth freedom operators, exercising their fifth freedom rights aggressively, in a manner consistent with the letter of
the bilateral agreement, but contrary to the desires and expectations of the competing airline. Sanctions have
included predatory competitive prices, limiting fifth freedom rights in subsequent negotiations, and outright
renunciation of the agreement. Even without such hurdles, many fifth freedom operations face operational problems
and scheduling compromises.

60
It is often difficult to identify a before and after period for liberalization. The
impacts often need many years to work themselves out, and the more drastic the
change; the longer will be the adjustment period.

Those specific countries which have chosen to liberalize their bilateral agreements are
by definition the primary source of before and after case histories. This will create a
selection bias, in that those instances chosen will involve nations with similar aviation
policies. A large proportion of the case histories must involve the United States or the
European Union.

Only the time series/case history approach can effectively consider the countless
behavioral elements that enter into a relationship. It can trace out many dynamic
elements, and the subtleties of the pre-liberalization regime, in a manner that is outside
the analytical straight-jacket of a formal model. A case history can best illustrate how
liberalization takes place and show the pathway of transformation. However, its findings
cannot be readily generalized.

The time series or case history approach requires a detailed analysis of traffic and
socioeconomic variables. It must also consider qualitative issues such as international
relationships, the often-convoluted processes of airline industry restructuring, and social
issues unique to the specific countries. However, the usefulness of this approach for
projecting results to any arbitrary country-pair is problematic.

2. THE CROSS-SECTIONAL APPROACH


The cross-sectional approach involves analyzing a minimum of two, but ideally many
thousands, of country-pair aviation relationships at the same point in time. Rather than
hold certain base conditions constant for a few narrowly defined variables, it assumes that
a particular relationship between traffic, the extent of liberalization, and socioeconomic
conditions applies to every market. Each country pair will display unique traffic volumes,
socioeconomic variables, airline industry conditions, and degrees of liberalization in the air
service agreements. Through correcting for variations in economic activity and other
extraneous factors, this approach seeks to explain variations in the passenger and air
freight traffic between different country-pairs to variations in their bilateral agreements. In
theory, this method should isolate the separate impacts of route definitions,
single/multiple designations, pricing controls, the presence or absence of fifth freedom
permissions, and other attributes of air service agreements. Through using a very large
sample, involving all regions of the world, nations in all stages of development, and
countries with a wide range of approaches to international aviation, the process should, in
theory, yield a robust estimate of the impacts for any arbitrary country-pair.

The cross-sectional approach shares many of the pitfalls of the case history/time series
methodology. Each nation-pair in the sample faces a unique set of economic and political
conditions. The case history approach addresses each aspect individually and corrects for
individual variations as they arise. The cross-sectional methodology, using a sufficiently

61
large sample, assumes random variations in the every element in the sample, and draws
patterns from a statistical process.

The cross-sectional approach poses analytical issues of its own:

An air service agreement is only one, albeit a very important, facet of a broader
aviation regime. Bilateral agreements can include literally hundreds of clauses and
restrictions, including issues that cannot be expressed in any simple mathematical
manner. The cross-sectional methodology, nevertheless, attempts to encapsulate all
aspects of the relationship into a few formal and well-structured variables.

A bilateral agreement reflects every aspect of the aviation policies of each member of
the country-pair. The attributes of any agreement are likely to be similar in their
reliance on economic forces. This creates estimation problems.48

It does not address the factor of time. The amount of time needed for a market to
respond to changes in the underlying regulatory approach can range from a few
months to several decades.

This approach, like any statistical research, is limited by the availability, quality and
timeliness of the data. Of the more than 40,000 nation-pairs, data could be obtained
for only 1,400. Those countries with good data appear most frequently in the sample.
Since the quality of data may be correlated with levels of national wealth and economic
activity, this process is systematically biased to evaluating particular markets. This
selection bias was moderated by using a large and varied sample.

The process includes country-pair markets of all sizes. Many statistical processes could
place unduly large weights on the largest markets. Several adjustments successfully
minimized this problem.

The cross-sectional approach ignores interactions between country-pairs. A strong


airline response to the permissive agreements between nations A and B may divert
traffic from the more protective A-C and A-D country-pairs. This situation will result in
an inaccurate reporting of country-pair flow data, in which the impacts of the liberal A-
B agreement are over-estimated by a model, and the adverse consequences of the
other restrictive agreements are similarly exaggerated.

48
This problem means that certain components of the sample data may have strong interrelationships. For example,
a country-pair with a permissive capacity clause in its bilateral agreement will likely also have permissive pricing
rules. Should this country-pair exchange more traffic than is commensurate with the GDP, community of interest
and other factors, it will be difficult to determine if the variation reflects the capacity provision or the pricing
formula. The more bilateral agreements that share these associations of their different attributes, the greater will be
the difficulty in determining the incremental impact of each attribute. Very modest changes in the underlying
structure can cause drastic changes in the outcome. At the extreme case, where all agreements reflect the
associations between one or more attributes, the model will fail to provide any estimates of any determinant. The
model will be attempting to obtain more information than the sample can provide. This is a common statistical
problem called multicollinearity, which can result from many factors, ranging from extreme operator carelessness
to just bad luck.

62
The cross-sectional method does not consider the lengthy history of a relationship. The
traffic flows in a recently liberalized market may not have expanded to their equilibrium
levels. Through failing to consider the time dimension, the cross-sectional approach
would understate the economics of liberalization. Furthermore, this class of models
does not consider adjustment times or mechanisms, but merely assumes that every
data point displays an equilibrium between traffic, the economy, and the applicable air
service agreement.

3. THE APPROACHES SELECTED


Both the time series/case history and the cross-sectional methods offer distinct
advantages. Their complementary relationship calls for both to serve as key foundations
for this study. The case histories highlight recent and very dissimilar instances of air
service liberalization. They consider the dynamics of the process and the path of
transition. The cross-sectional model offers a formal and rigorous framework that can be
universally applied and that can produce extrapolations for any arbitrary country-pair, but
cannot by definition account for individual nuances. The next section summarizes the
development of a formal cross-sectional econometric model of the impact of the bilateral
regime on air traffic.

C. THE CROSS-SECTIONAL MODEL FOR PASSENGER TRAFFIC

The model expresses the air traffic between any particular country-pair as depending on a
vector of geographical, socioeconomic and regulatory variables. The model considers each
country-pair as an independent entity; its traffic will not be affected by changes in other
country-pairs. Furthermore, events in other economic sectors, such as new consumption
opportunities that may compete with air travel, will not affect traffic in any manner.

Each data point consists of one country-pair. The dependent variable consists of the yearly
two-way origin-destination traffic between the country-pair. The model views passenger
traffic as a function of several socioeconomic and geographic variables, and the chosen
attributes of the relevant bilateral air service agreement.

1. DATA SOURCES
The endogenous variable, country-pair origin-destination traffic was provided by IATA.49
The World Bank, International Monetary Fund, World Tourism Organization, Population
Reference Bureau and the United States Central Intelligence Agency provided data on
exogenous variables from several publications and websites.

49
Source: International Air Transport Association, Passenger Forecast 2005-2009, reference 9261-05, Geneva,
2005.

63
2. SOCIOECONOMIC AND GEOGRAPHICAL VARIABLES

a) Gross Domestic Product


Gross Domestic Product (GDP), calculated from the Purchasing Power Parity
method measures the total magnitude of economic activity in any nation. The
specification assumes that changes in the GDP of each country in the country-pair
will have identical influences in the level of traffic.

The GDP term proved the most important exogenous variable in terms of
significance and explanatory power. Its presence in a model sometimes rendered
other variables superfluous.

b) Moment of Inertia
Travelers must decide between domestic and international flights. A nation with
many opportunities for domestic air travel, other factors being held constant, will
generate fewer international passengers than one with few domestic markets. For
example, a resident of India can fly to a very large range of domestic destinations,
offering a huge range of natural and cultural diversity. In contrast, a Singapore
national can only fly to a foreign destination. It might naively be concluded than
Singapore nationals have a greater behavioral propensity for foreign travel than
those of India. Rather, they merely lack domestic destinations to fly to. Several
variables were tested to measure domestic travel opportunities, including total
area, population, and the number of airports with commercial services in each
country. The best measure reflected both the area and the geometrical shape of
each country.

In physics, the moment of inertia of an object represents its resistance to any


rotational force. This quantity reflects the weight of the object and its shape. A
long, narrow shape will have a larger moment of inertia around its center than a
square shape of the same mass. The moment of inertia for a country is
calculated from its area, the length of the coastline and the length of its land
border. The model views each country as a rectangle, calculates the two
dimensions, and applies the definition of the moment inertia50 about the geometric
center. A large moment of inertia could indicate either or both a large area and an
elongated shape, in which there are many cities, and distances between them are
large. It could also indicate a nation with many islands. A small moment of inertia
could result from a small area, or a square shape. In such a situation, distances
between cities will be short, and there will be few domestic city-pairs requiring
scheduled air service. For example, Saudi Arabias area is 2.1 percent greater than

50
For a rectangle about its geometric center, the moment of inertia is one-twelfth the mass times the sum of the
two dimensions squared. In this case, the mass is defined as unity. The two dimensions are calculated from the
area, the length of the land boundary and the length of the coastline.

64
Indonesia, but the latters moment of inertia is over 96 times larger. As a huge
archipelago, Indonesia offers correspondingly more opportunities for domestic air
travel. Similarly, Chiles elongated shape makes its moment of inertia 175 percent
greater than that of Myanmar, although its area is only 11.6 percent larger.

The model uses the product of each countrys Moment of Inertia to represent
opportunities for domestic travel, which may compete with an international
service.

c) Distance between the Countries


This variable represents the distance between the two countries. Each nation is
represented as a single point, usually its primary international airport (e.g.
Copenhagen for Denmark). St. Louis, close to the geometric and population
center, represented the location of the United States.

d) Flows of Services
Unlike goods, services are consumed at the same time and place as they are
produced. They cannot usually be stored in inventory. Service activities include
insurance, financial assistance, medical services, management and consulting.
Since they usually require a close interaction between the seller and the
consumer, the sale of services is an important determinant of the demand for
travel.

Including disaggregated services trade data for each potential country-pair would
be costly and awkward. The model therefore uses a gravity-type relationship
between each nations services trade with all countries to define a country-pair
propensity. The Service Flows term for the country A-B was expressed as:

Exports of Services by Country A * Imports of Services by Country B


+
Exports of Services by Country B * Imports of Services by Country A

No statistical advantages resulted from applying a similar formulation to the


Commercial Service, Finance/Insurance or Computer/Communications separately, and
incorporating each of the now three predetermined variables individually into the
model.

e) Intervening Opportunities
The traffic between any country-pair will be less if passengers could choose from
other, closer destinations. For example, Australian residents will view New Zealand

65
as easier as and cheaper to reach than the United Kingdom. This proximity will
correspond to a lower demand among Australians for air travel on the Australia-
United Kingdom route. Similarly, individuals and businesses of the United Kingdom
may view Canada as a partial substitute for Australia. This would reduce the
volume of Australia-destined traffic originating in the United Kingdom.

The passenger model uses an Intervening Opportunity quantity as a determinant


of country-pair traffic. For each country in a country-pair, the model calculates the
sum of the GDPs of every country that is 10 percent or less distant than the other
nation in the country-pair. The resulting sum measures the size of closer
opportunities. The product of the Intervening Opportunity term for both nations in
a country-pair proved to be a useful predictor of country-pair traffic, and displayed
the expected negative sign.

3. VARIABLES PERTAINING TO BILATERAL AGREEMENTS


The model represents the attributes of bilateral agreements by five discrete (0 or
1) variables. A 1 designates a restrictive provision. The degree to which each
attribute constrains passenger traffic depends on the specific situation. Multiplying
the five 0-1 measures by geographical or socioeconomic variables created a new
set of exogenous variables that measured the relevance of each bilateral
constraint to the country-pair in question. A description of each of the five
attributes follows.

a) Permitted Number of Airline Designations


Bilateral agreements usually specify the number of airlines permitted to fly any
route between the two countries. A 0 denotes a dual or multiple designation; a
1 otherwise. This digit is then multiplied by the distance between the two
countries. A country-pair can only benefit from a multiple designation if one or
both countries have more than one airline fit, willing and able to operate the route.
Furthermore, each such country must be willing to allow its own airlines to
compete.

An airline seeking to operate long distance services must usually use wide body
aircraft. It will require a network of feeder services using smaller aircraft. In
contrast, many short haul services use much small aircraft, and can serve strictly
point-to-point markets. The airline operating long haul services requires very
substantial physical and financial resources. Comparatively few countries have
more than one airline operating long distance services. Many are more
conservative in allowing competition between their airlines on intercontinental
routes, compared to shorter and highly fragmented regional markets. A
single-designation rule would therefore be more onerous to short distance services
than to longer flights.

66
b) Capacity Controls
Many regulators consider capacity controls as particularly inimical to market
growth, and a key trait of a restrictive agreement. Sometimes, the limits are
written directly in the agreements. Lengthy negotiations are often necessary to
increase the limits. In other instances, such as Bermuda agreements, the
capacities are subject to a regular process of consultation. In either case, the
airlines flying between the two nations have many opportunities to curb capacity
growth and maintain high fares.

Sometimes the capacity controls are not binding; more flights are permitted than
the airlines actually operate.51 The nations negotiating bilateral agreements will
forecast demand, and negotiate formal limits accordingly. While such agreements
reflect a very strong interventionist approach, they do not necessarily constrain
traffic. In many conditions, passengers traveling between two countries can transit
via third nations. This can mitigate the impacts of the capacity controls governing
the direct routing. It can force the negotiators to develop more liberal capacity
provisions than otherwise.

Two variables were employed to model the impact of capacity controls. The first
was a 1 if capacity was fully predetermined by the agreement (which
corresponds to the most inflexible form of capacity clause), and zero otherwise. A
second 1-0 vector applied if a Bermuda-type clause was in force. Both dummy
variables were multiplied by the GDP, reflecting a hypothesis that capacity controls
become proportionately more detrimental to competition as the size of the market
grows.

c) Pricing
This variable is assigned a 0 if the bilateral includes a double-disapproval (a
proposed fare would be permitted unless both nations veto it) clause to signify the
most permissive form of pricing52 enforcement. A 1 indicates another regime,
such as country-of-origin, zone-of-reasonableness or single disapproval pricing.
The resultant quantity was then modified by the product of the per capita GDPs of
both countries. This reflected the belief that countries with a large per capita GDP
would be most likely to generate large volumes of leisure travelers. They would be
especially affected by any price rigidities. Furthermore, airlines are most likely to
offer incentive fares on routes with considerable leisure traffic. A restrictive pricing

51
Indeed, capacity controls often lead to more capacity than is appropriate for a market. Sometimes, several airlines
of one nation must compete for scarce route authority. Often, a use it or lose it rule encourages an airline to
continue an unprofitable service, rather than discontinue the route and lose the authority permanently. Liberal
bilaterals, by granting carriers freedom to enter a market, also enhance freedom of exit. They help the industry
react to shocks such as the traffic downturn following the 9-11 terrorist attacks on Washington and New York.
52
Even this provision is stricter than corresponding restrictions in other forms of trade. Anti-dumping rules are often
difficult to enforce.

67
regime, which limits their flexibility, would be a proportionately large obstacle to
growth in affluent country pairs.

Specifications that considered the portion of GDP accounted for by manufacturing,


primary production or services, the degree of income inequality as measured by
the gini coefficient, and the per capita income or GDP failed to produce a model
having desirable statistical properties, or intuitively plausible results.

d) Fifth Freedom Rights


A 1 indicates the absence of any fifth freedom rights in the bilateral. A 0
depicts an agreement with such provisions. The data did not permit a more precise
delineation of fifth freedom rights, such as between intermediate and beyond
rights.53 Even this definition greatly oversimplifies such rights; the economic value
of seemingly identical provisions varies widely. As indicated earlier, there can be a
large grey area in the interpretation of these rights in any situation.

Fifth freedom rights can be most valuable for long haul services, for which
intermediate stops may be technically necessary. An ability to top off a long
distance flight with incremental short haul revenue, or serve a minor center as
part of a longer flight to a more significant destination may be necessary for a
profitable route. These factors suggest that a fifth freedom provisions may be
more important to nation-pairs that are relatively distant. Furthermore, other
significant markets should occur either in close proximity to the great circle flight
path between the two nations (for intermediate fifths) or reasonably close to either
nation. The 0-1 variable is therefore multiplied by the product of the intervening
destinations variable to measure the significance of fifth freedom services for each
country-pair observation.

This approach proved statistically meaningful and plausible. The Intervening


Opportunities variable depicts intermediate fifth freedom rights as less valuable to
flights operating over remote areas or long, trans oceanic sectors (e.g. Auckland-
Buenos Aires) than those overflying large and populous land masses. It also
assigns a value to relatively short beyond rights. Its major weakness is an
inability to capture very long beyond fifth freedom sectors, and thereby to
under-estimate the importance of fifth freedom rights on a small number of
specific flights. For example, Air India exercises London-New York traffic rights on
its India-United States services. This capability allows it to offer better services
between India and the United Kingdom. The model does not, however, consider
the United States as an intervening opportunity or a potential beyond market

53
Consider the bilateral agreement between nations A and B. An intermediate fifth freedom right would allow
an airline of nation A to operate a flight that departs from nation A, stops at a point in nation C, and continues on to
a point in nation B, carrying C-B local traffic in the process. A beyond right would allow an A-B-C routing, again
with B-C local traffic rights.

68
for India-Britain. Any metric that adequately models the relatively few long haul
beyond fifth freedom sectors would grossly overestimate the value of fifth
freedom rights for most country-pairs.

e) Named Points
Some bilateral agreements limit services to a very few rigidly defined destinations;
others, following a more liberal approach, allow services to any operationally
feasible combination. In many situations, bilateral agreements will stipulate a fixed
number of roving points, for which each nation can choose the precise
destinations at a later date.

A very flexible definition of permissible routes is most conducive to competition


when it involves nations with large areas and many potential destinations. A long,
linear nation, or one with many distinct land masses, would likely offer several
attractive cities able to support international service. Mean travel times from the
periphery to a single, central international airport would be longer than for a
square-shaped nation of similar area.

This variable was assigned a value of zero for country-pairs with broad route
definitions. Those observations with specific point restrictions were assigned a
value equal to the product of the moments of inertia of the two participating
countries.

f) Other Attributes
Air service agreements can include other provisions, including mandatory pooling,
royalty payments, restrictions on airline designations, traffic thresholds to trigger
certain actions such as allowing further designations, flight timing restrictions,
provisions for charter traffic, access to runway and gate slots at busy airports,
currency repatriation and facilitation and others. The complexity and diversity of
these provisions can only be examined effectively by a comprehensive analysis of
an individual country-pair.

4. OTHER SPECIFICATIONS
Several exogenous variables were tested and subsequently rejected. Most proved
statistically significant in certain specifications. This often reflected the large sample size
(up to 1,400 data points), which captured even trivial influences. Many did not justify the
additional complexity, or exacerbated multicollinearity concerns. Those variables
ultimately selected provided the best combination of statistical power, simplicity and policy
relevance. Exogenous variables examined and ultimately rejected include:

69
a) Tourism Demands and Market Drivers
Traffic on some country-pair markets consists primarily of tourists. Traffic flows
are far larger than what would be commensurate with the GDP of the destination
country. Every country-pair in the model was examined. In some instances, the
proportion of employment accounted for by the tourism economy was considerably
larger in one country than in the other. In such instances, traffic would be driven
primarily by the GDP of the source country. A second factor would be the
availability of other tourist-oriented economies located closer to the source
country. For example, travel on the United Kingdom-Maldives country-pair would
consist primarily of British vacationers. It would be highly sensitive to economic
conditions in Great Britain. The model would also consider terms measuring the
availability of closer, competing destinations, such as Greece or Spain.

These criteria were incorporated into the simpler specifications discussed earlier.
While the regressions generated significant coefficients of the expected signs, their
magnitude made little difference in the outcome of the model. The sheer size of
the sample, up to 1,400 country-pairs, made it prone to pick out statistically
significant but ultimately trivial differences. The tourist activity column vectors
contained many 0 elements, exacerbating the problems of multicollinearity. These
difficulties were especially severe when the tourism variables were included along
with those examining bilateral agreements. The latter set of variables was included
because of its more fruitful policy implications.

b) Cultural Affinities
Two nations that together were once part of the same colonial empire may share a
common culture and political institutions. They may also have preferential trade or
immigration policies. These common factors would encourage air travel between
the two countries. Similarly, a common language or religion would, ceteris paribus,
be expected to exchange more passenger traffic than otherwise. The extreme
instance involves Islamic nations and the Haj. The religions were highly
aggregated, so that Protestant or Roman Catholic faiths were aggregated into a
single Christian category, while Shia and Sunni Moslems were considered
Islamic. The model assigned a 1 to nation-pairs sharing one or more religions,
and 0 otherwise.

The affinities specifications proved significant, with the expected positive signs,
when superimposed on a model containing only socioeconomic variables. However,
the estimation process broke down when the metrics for bilateral agreements were
also included. The sparse matrices, and multicollinearity, greatly reduced the
usefulness of the estimates.

70
c) Tourism and Travel Industry Exogenous Variables
Several variables measuring the size of each nations tourism and travel industries
were examined. They were generally significant, but not remarkably so. In a
model of country-pair traffic flows, these variables should be considered as
endogenous, thereby introducing fundamental issues with the estimation
algorithms.

d) Quality of Service
The large database of schedules maintained by the Official Airline Guide provides a
mechanism to develop measures of the quality of air service between any country
pair. It should reflect aggregate capacity, frequencies, number of city-pairs
served, number of airlines offering competitive services, and competing routes
that use connections in third countries. Such a measure is so strongly correlated
with the level of traffic that it will overwhelm the exogenous variables. It could
serve as an instrument for demand.

e) Price
The effective price of air travel is a critical determinant of travel demand.
Liberalization affects price in many ways, through creating greater flexibility,
depriving a nation of the ability to veto prices charged by its airlines competitors,
forcing airlines to address cost issues, permitting capacity growth, encouraging
new entrants, and forcing the incumbent airlines to address the threat of new
entrants. A two-step model could relate liberalization to reduced prices, and use
demand elasticities to calculate the ensuing growth of traffic. It could also model
the consequences of international service by low cost carriers. A lack of high
quality information on average fares by country pair precluded any detailed
consideration of price influences.

f) Other Variables
Other candidate determinants of travel included total merchandise imports and
exports, the distribution of income, as measured by the gini coefficient, population
and the degree of urbanization. None of these variables demonstrated any
significant value as exogenous variables, once the GDPs had been included in the
model. The number of airports with commercial service was tested as an alternate
variable to the moment of inertia. It was weaker, and could pose problems for
updating the data matrices in the future.

71
D. ESTIMATION OF PASSENGER MODEL

1. REGRESSION ANALYSIS
The preliminary estimation process used an ordinary least squares algorithm on a double-
log specification. This reflects the assumption that many of the processes being modeled
are multiplicative. For example, a restrictive bilateral would cause a greater absolute loss
of traffic in a large market than in a small one.

As is common with many cross-sectional models, the preliminary specification showed


problems with heteroscedasticity,54 as determined by a significant Goldfeldt Quandt
statistic. A general least squares procedure, using the GDP variable as a weighting factor,
produced the estimates shown on Table IV-1.

Table IV-1: Regression Results for the Passenger Model

95% Confidence
Interval

Variable Coefficients Standard T Lower Upper


Error Statistic
Intercept -0.42345 0.277463 -1.52615 -0.96809 0.121191
1.
Single Designation -0.02101 0.011204 -1.87533 -0.043 0.000981
2.
Predetermined Capacity -0.03687 0.01397 -2.63921 -0.06429 -0.00945
3.
Bermuda Capacity -0.02578 0.014781 -1.74384 -0.05479 0.003239
4.
Single Disapproval Pricing -0.03629 0.015276 -2.37542 -0.06627 -0.0063
5.
Fifth Freedoms -0.00036 0.003121 -0.11385 -0.00648 0.00577
6.
Authorized Points -0.05866 0.027313 -2.14783 -0.11228 -0.00505
7.
GDP Product 0.240543 0.040627 5.920825 0.160796 0.32029
8.
Commercial Flows 0.14279 0.033162 4.305882 0.077696 0.207884
9.
Intervening Opportunities -0.05739 0.005125 -11.1966 -0.06745 -0.04733
10.

54
Any regression model seeks to explain each observation of a y variable, in this case traffic, in terms of one or
more x variables. There is usually an unexplained quantity for each data point; the disparity between the predicted
value and the actual observation. Over the entire sample, this residual averages to zero. Sometimes, this residual
varies according to some pattern, rather than being totally random. For example, the residuals for country-pairs
with a large GDP may average zero, but could consist of many large negative and positive values. Observations with
a small GDP could include only small positive and negative deviations. This property, called heteroscedasticity,
means that the ordinary linear regression, while producing estimates with many desirable properties, has failed to
extract all of the information available from the sample. Furthermore, some of statistical tests lose validity. Various
estimation techniques can overcome these problems and produce new estimates of the model with superior
properties. While statistically significant, the magnitude of the heteroscedasticity was smaller than expected.

72
Multiple R 0.671431

R Square 0.45082

Adjusted R Square 0.444641

Standard Error 0.099722

Observations 810

Source Degrees of Freedom Sum of Squares Mean Square F Statistic


9 6.530724 0.725636 72.96844
Regression
800 7.955614 0.009945
Residual
809 14.48634
Total

The regression shows a reasonable fit, and the signs are consistent with expectations.
The variables 2-7 investigate the impact of selected constraints of bilateral agreements on
traffic volumes. The uniformly negative signs are evidence that the artificial
constraints posed by bilateral air service agreements constrain the growth of
traffic. Furthermore, these obstacles operate, not only between specific well-
studied country-pairs such as between the United States and the United
Kingdom, but also in a huge variety of markets, involving countries of all sizes,
stages of economic development, and political systems, in every part of the
world.

These results therefore support the hypothesis that restrictive bilateral agreements
constrain traffic development. They lead to the rejection of the null hypothesis, that
restrictive bilateral agreements have little impact on traffic.

The fifth freedom variable was negative, indicating that bilateral agreements with fifth
freedom provisions could promote traffic development in the corresponding third and
fourth freedom markets. However, the coefficient was not significantly different from zero.
The model therefore does not support a liberal policy on fifth freedom rights. Rather, it is
far more important to get it right on the third and fourth freedom clauses.

This negative finding on fifth freedom rights is not remarkable. As traffic grows, more
country-pairs can support nonstop turnaround services. Aircraft manufacturers are
developing ultra-long range aircraft, smaller than those that have traditionally operated on

73
long distance services, to offer nonstop flights on long and thin routes. The market
fragmentation could see nonstop international services to city-pairs that once relied on
intermediate hubs. As traffic grows, third and fourth freedom airlines can use their market
strengths to displace fifth freedom operators. For example, Malaysia Airlines shifted its
Newark-Dubai-Kuala Lumpur flight to a routing via Stockholm when Emirates commenced
nonstop New York-Dubai flights in 2005.

The coefficient for GDP is much lower than that estimated by other studies. Most research
concludes that traffic grows faster than GDP, as indicated by a coefficient, representing
the elasticity that exceeds unity. Although the estimated coefficient refers to the product
of the respective GDPs, and should therefore be doubled to provide a basis of comparison,
the resulting elasticity is still much less than generally accepted values. Most studies trace
the growth of one nations air travel demand and its GDP over time. The cross-sectional
approach captures very different economies at a single point of time. The differences in
GDP and air travel demands between a small, undeveloped nation and a large,
industrialized economy are far more profound than the differences relating to one
economy and a before-after interval of a few years. The resulting elasticities are not
comparable.

The diversity of country-pairs in the model generated these disparities. A simplified


model, investigating a greatly reduced data set of countries with large GDPs, high volume
traffic flows, and an established history of direct services, and exploiting the a-priori
knowledge available in more traditional approaches, yielded estimates of elasticities that
are fully consistent with prevailing time series studies and elasticity measures.

2. ADJUSTMENTS
The regression model described in the previous section relates traffic volumes between
any arbitrary county-pair to socioeconomic parameters and the bilateral air services
agreement. Variables (2)-(7) together define the incremental traffic growth resulting from
total or partial liberalization of the air services agreement.

The model demonstrates the importance of liberalization to the growth of international air
traffic. However, the idealized model poses several challenges to any attempts to apply it
to specific situations. The model uses a double-logarithm specification. This reflects the
interdependency of the exogenous variables. However, such models tend to dampen any
disparities between predicted and actual traffic observations. Even a small residual of such
a model can result in a very large disparity when the coefficients appear as exponents.

The coefficient corresponding to the Bermuda capacity control was not statistically
significant. Although lower than the predetermined capacity coefficient, it is believed that
the model may still overstate the traffic stimulus arising from the abolition of Bermuda
controls. The predictive model therefore assumes that a Bermuda clause is half as
detrimental as a capacity stipulated directly in the bilateral agreement. The raw model

74
also suggested that a very large stimulus of 75 percent or more would result from
liberalizing price clauses (Variable 5). The results, when applied to several country-pairs,
appeared excessive. A separate analysis, isolating this term alone, suggested a stimulus of
4.1 percent was more appropriate. This finding was superimposed on the predictive model,
although it is fully recognized that this ad hoc measure is a methodological compromise.

The predictive model reflects a view that restrictive bilateral agreements, whatever their
economic justification, are the global status quo. It is usually thought that any unjustified
change in the status quo is a more serious error than mistakenly retaining the current
circumstances. This in turn means that an error in over-estimating the consequences of
liberalization is more serious than any under-estimation. The model therefore tests each
prediction. Should the predicted stimulus exceed a particular critical value, the stimulus is
reduced to that particular value. No such correction applies if the model appears to under-
predict the consequences of liberalizing a particular country pair. Furthermore, a grand
limit capped the total growth resulting from a full liberalization.

The limits were estimated by taking a sample of 600 country-pairs in various stages of
liberalization. Each attribute of the relevant bilateral agreements was examined in turn,
and subject to a step-by-step liberalization. The model calculated the conditional
expectations of traffic resulting from each perturbation of the bilateral for each
observation, generating a series of calculated stimuli. For each attribute in the bilateral, a
maximum limit on the traffic gain from an incremental liberalization was calculated using
Chebyshevs Inequality.55 The process yielded, for each attribute, and for a total
liberalization, a level of stimulation that would be exceeded by only 10 percent of the
observations. To eliminate the risks of over-estimating the stimulus from liberalization, the
model superimposed the limits shown in Table IV-2 on any extrapolation produced by by
the regression model.

Table IV-2: Maximum Stimulation Limits

Maximum Permissible Traffic


Liberalization Measure
Growth
Single to Multiple Designation 50.7%
Predetermined Capacity to Open Capacity 25.0%
Bermuda Capacity Control to Open Capacity 17.8%
Single Refusal to Double Refusal Pricing 4.1%
Including Fifth Freedom Rights 8.8%
Named Point Route Annexes to Open Routes 97.3%
Fully Restrictive to Fully Liberal 166.4%

55
Chebyshevs Inequality describes very broad characteristics that govern any statistical population. It is
distribution free in that it does not require any prior knowledge of the population, except that it have a mean and
variance.

75
The procedure also generated lower limits to stimulation factors. The simulation did not
enforce the lower limits, in keeping with its conservative stance on the air service issue.

Liberalization is a necessary but not a sufficient condition for traffic growth. No new
services can result if there is no underlying demand to support them. The model therefore
examines the air services already operating between the country-pair in question. If any
such flights already operate, it is assumed that capacity can expand to accommodate
demand. If no such flights exist, the algorithm determines the aircraft most appropriate
for a route of that length.56 If the traffic available is insufficient to support an arbitrary
three weekly return flights with a 70 percent passenger load factor, the algorithm
considers that no service is feasible. The model then examines the bilateral agreement to
ascertain if fifth freedom rights are available. If so, it allocates half of the capacity to the
fifth freedom market, and chooses an aircraft appropriate to half of the distance between
the two countries. Only if the country-pair can meet the 70 percent load factor and
frequency requirements will it obtain direct service. Otherwise, the liberalization is
considered moot; no traffic increase could occur.

E. ESTIMATION OF CARGO MODEL

1. INTRODUCTION THE MODELING CHALLENGE


Liberalization can be especially important to air cargo. Indeed, the rise of integrated
carriers such as UPS and FedEx is itself a direct consequence of airline deregulation. Many
bilateral agreements have liberalized provisions for all-cargo services, including seventh
freedom57 rights. All-cargo services are sometimes used as a means to test the
consequences of liberalization before extending the process to include passenger services.

Air freight58 poses a particular challenge. It is very heterogeneous, and includes


documents, machinery, foodstuffs, live animals and literally anything else that people wish
to ship. Shipments range in size from single-page documents to huge movements
requiring charter of several wide body all-cargo aircraft. The institutions vary. The
integrated carriers offer a seamless, door-to-door service. They perform all functions in-
house: sales development, pickup and delivery, tracing, carrying the goods by a variety of
modes, customs clearance, insurance, and often, warehousing, inventory control and other

56
There is a rough relationship between aircraft size and flight distance. Large, twin-aisle aircraft operate most
intercontinental routes. Smaller aircraft customarily serve shorter routes. An analysis of Official Airline Guide
worldwide schedules for March, 2006 identified the type of aircraft appropriate for each stage length. The
economically meaningful maximum range of an aircraft was defined as that distance exceeded by 10 percent of the
flights.
57
See Glossary, Appendix B.
58
Air cargo is usually viewed as any kind of movement of merchandise by air. It includes air mail, which is
tendered by a government-owned postal service, usually under a long term contract. Air freight is a narrower
definition, that excludes air mail. It consists of revenue-generating traffic tendered by entities other than the post
office. This Report will use air freight in this narrower sense. Any mention of cargo will include air mail.

76
aspects of the management of a client companys supply chain. Flights are scheduled to
meet the needs of air freight, and the carrier holds all route authorities.

In contrast, traditional airlines regard air freight largely as incidental revenue, using the
otherwise empty belly space of passenger flights. Many entities, forwarders, customs
brokers, truckers, pickup and delivery operators, sufferance warehouses, etc. will
participate in a single shipment. Besides its heterogeneity and its by-product status, air
freight differs from passenger traffic because goods usually travel in only one direction.
Services on an individual country pair may face a chronic traffic imbalance, with loads in
one direction continually exceeding the other. Sometimes, the airline will carry more traffic
in the weak direction, accommodating dense, low yield items such as perishables. The
airlines commitment to cargo varies widely, with Asian and European airlines usually
showing the greatest interest. Carriers usually design their networks and schedules for
passengers. The air freight industry makes extensive use of supplemental trucking
services, and published statistics usually provide limited information about the
fundamentals of a particular traffic flow.

These complications complicate development of a simple, accurate air freight model that
can apply to any country-pair. Rather, an intensive, market-by-market approach is
essential. This approach considers the underlying economic fundamentals, the industrial
base, the available air services, directional balances, cargo yields, the passenger market,
intermodality and the identities and strategies of specific shippers and consolidators.
Instead, the model developed for this report has the less ambitious target of estimating
the changes in air freight and their economic impacts as consequences of the
liberalization of passenger services. It does not consider how liberalization could affect
the supply and demand for capacity on all cargo aircraft.

The air freight model therefore does not attempt to relate traffic directly to liberalized air
service agreements.59 Rather, it calculates the additional passengers resulting from the
liberalization, the incremental passenger capacity that they will need, and the new belly
cargo capacity that would be forthcoming. The model then assumes that airlines will
manage their cargo yields so as to fill the new capacity. This capacity-driven implies that
belly air freight traffic will grow at the same rate as passenger volumes. However, shipper
demands for cargo capacity respond to a radically different series of stimuli. This
simplification therefore does not explain growth in pure freighter services.

59
Directional imbalances are of such profound importance to the air freight industry that traffic in each direction of a
country-pair must be considered as a separate observation, as a minimum. The IATA publications from which the
country-pair traffic was assembled did not provide directional breakdowns. They showed directions by nation to
certain geographical regions, and consolidated flows by nation pair. It was necessary to estimate directional
breakdowns through an iterative, multi-step procedure. Nevertheless, many country-pair data, especially for
important trading nations, had to be allocated into directional components. This data shortcoming alone limits to
model to belly air freight on combination carriers.

77
2. THE AIR FREIGHT MODEL
The air freight model is similar to the passenger traffic model. Both use a gravity-type
double-logarithmic specification. The air freight model considers each direction of a
country-pair as an independent observation, and benefits from a correspondingly large
number of data points.

The GDPs of both the importing and the exporting countries were important determinants
of demand. The coefficient of the exporting country was both slightly larger and more
significant than that of the importer. The presence of closer opportunities, whether as
markets for the exporter or competing sources of supply for the importer, was significant
and negative. Distance was a positive and significant factor. A large distance between the
exporting and the importing country implies higher freight rates, but also suggests major
advantages for air freight over surface modes.

During the process of developing the model, other variables emerged as candidates. These
included exports and imports of food and merchandise, the total manufacturing GDP,
degree of urbanization, and relationships between exchange rates and purchasing power
parity measures (as a proxy for the price competitiveness of exports). None of the
augmented specifications generated enough explanatory benefits to justify the added
complexity.

Table IV-3 summarizes the results of the air freight regression. The ordinary least squares
algorithm provided significant estimates. Despite the size and diversity of the cross-
sectional sample, heteroscedasticity did not pose a complication. Although many of the
exogenous variables were correlated, multicollinearity did not affect the estimation
process.

Table IV-3: Ordinary Least Squares Estimation of the Air Freight Model

95% Confidence Interval


Standard
Variable Coefficients Error T Statistic Lower Upper
1. Intercept -2.81649 0.195981 -14.3713 2.71E-44 -3.20088
2. Distance 0.580981 0.053722 10.81455 2.07E-26 0.475613
3. GDP Exporting Country 0.501725 0.023073 21.74534 9.67E-93 0.456471
4. Intervening Opportunities Exporter -0.11546 0.022093 -5.2261 1.94E-07 -0.15879
5. GDP Importing Country 0.454237 0.022648 20.05604 1.65E-80 0.409815
6. Intervening Opportunities - Importer -0.09314 0.021884 -4.25626 2.19E-05 -0.13607

78
Multiple R 0.612228

R Square 0.374823

Adjusted R Square 0.372991

Standard Error 0.712438

Observations 1712

Source Degrees of Freedom Sum of Squares Mean Square F Statistic

Regression 5 519.1548 103.831 204.5657

Residual 1706 865.9107 0.507568

Total 1711 1385.065

The overall model views the growth of air freight as an incidental result of the growth of
passenger traffic. Additional passenger traffic calls for additional flights, which in turn
increases the supply of belly capacity. On routes already served by direct flights, the
model assumes that the fleet mix will remain unchanged. The supply of air freight capacity
will then grow at the same rate as the passenger capacity and the passenger demands.

Section E-2 explained the rationale for assigning passenger capacity between newly
liberalized country-pairs, where no services presently exist. The process involves assigning
three weekly frequencies of a distinct type of aircraft to the country-pair. The total
incremental air freight volume is determined from the capacity of the aircraft, the
frequency, and whether capacity must be allocated to a fifth freedom destination.

The capacity of a particular passenger aircraft for air freight is often ambiguous. The
space relegated to cargo will depend on the number of passengers carried, their personal
luggage, and their preferences for checking items versus carry-ons. Passengers and
luggage often have a higher loading priority than air freight. Airlines may deliberately
leave belly space unused in order to accommodate any last-minute travelers. Low cost
carriers in particular have rapid station turnarounds, and their ground crews may be
unable to process air freight. Many shippers greatly prefer wide body aircraft with
container capabilities superior to bulk-loaded narrow body equipment. The quantity of air
freight that can be accommodated, usually measured by its weight, will depend on the
density of the items, their overall dimensions (large items often cannot be accommodated,
especially on narrow body aircraft), the type of materials (some hazardous materials
cannot be carried on passenger aircraft), whether the flight is carrying air mail, and the

79
airlines internal policies on loading priorities. Sometimes, the total gross takeoff weight
for long distance flights, or those forced to use short or contaminated runways, will be
limited for safety reasons, and the airline will forego all revenues for air freight. The
complications render published payload statistics of little relevance for defining the
effective air freight capacity of a flight.

The United States Department of Transportations T-100 Domestic and International


Segment reports offer a mechanism to calculate the effective air freight capacity of
different types of aircraft. A review of a full year of operating data for combination flights
generated a matrix of air freight weights carried per segment for each type of aircraft for
each month. The records for each type of aircraft aircraft were ranked in descending order
according to the average quantity of air freight carried. That segment corresponding to the
80 percentile defined the average effective air freight capacity for that aircraft type.
Capacities for aircraft missing from the T-100 report were estimated. For example, the TU-
154 was assumed to offer the same cargo capacity as the 727-200.

The air freight regression model plays a relatively limited role in this analysis. It defined
the extent to which shippers could use the additional capacity on passenger aircraft. It
also defined directional balances, with air freight volumes for the weak direction reduced
accordingly. The model is an essential part of the economic impact calculations, but does
not directly drive the links between liberalization and economic development.

F. ECONOMIC AND CATALYTIC IMPACTS

1. INTRODUCTION
While it is clear that air transportation has increased as economies have grown, it is also
recognized that air transportation is an important facilitator of general economic growth.
Not only is aviation a major industry in its own right, employing large numbers of highly
skilled workers, but more importantly it an essential input into the rapidly growing global
economy. For example:

Air transportation is a vital contributor to the development of tourism in many parts of


the world. This includes not just leisure tourism, but also business travel, conferences
and conventions.

Air transportation is used for the transportation of many high value goods to markets
around the world, and plays an essential part of just-in-time production and
distribution management.

It is an important input to many service and high-tech industries where interpersonal


communications are crucial. Air transportation can enhance the productivity of these
businesses.

80
The availability of air transportation can attract investment and new businesses. As
discussed in more detail below, company investment and location decisions are
strongly affected by transportation links, including air transportation. Countries and
regions with higher levels of air service are likely to attract more new businesses and
encourage existing businesses to expand.

This report examines three ways in which increased levels of air transportation can spur
economic development and employment:

Aviation Sector Impacts: economic activity in the aviation sector is related to the
servicing, management and maintenance of additional air services. This includes
activities at airlines, airports, air navigation and other businesses that support and
supply the aviation sector. The impact can spin-off into the wider economy
(multiplier effects) e.g., trucking, warehousing and logistics businesses facilitating
air cargo, or firms supplying food for catering on flights.

Tourism Sector Impacts: air service facilitates the arrival of larger numbers of
tourists to a region or country. This includes business as well as leisure tourists. The
spending of these tourists can support a wide range of industries: hotels, restaurants,
theatres, etc. Of course, air service also facilitates outbound tourism, which can be
viewed as reducing the amount of money spent in an economy. However, even
outbound tourism involves spending in the home economy: travel agents, taxis, etc.
Money spent by tourists flying abroad would not necessarily be spent on tourism at
home if there were no air service.

Catalytic Impacts: this includes the role of air transportation in facilitating growth
and productivity in the general economy by increased trade, business activity and
greater personal productivity.

Ultimately, through these three impacts, air transportation can contribute significantly to
rising GDP, higher disposable incomes and higher standards of living.

2. WHAT IS ECONOMIC IMPACT?


Economic impact is a measure of the spending and employment associated with a
business, a sector of the economy, a specific project, or a change in government policy or
regulation. Economic impact can be measured in various ways. Two of the most popular
ways to assess economic impact are the dollar value of output produced60 or the person
years (also known as full-time equivalents - FTEs) of employment generated. These

60
The two most common measures are economic output (Output) and gross domestic product (GDP). Economic
output roughly corresponds to the gross revenues of goods or services produced by an economic sector, while GDP
measures only value-added revenues. GDP removes the revenues to suppliers of intermediate goods and services
and only includes the contributions from labor and capital. Alternatively, economic output adds all revenues at each
stage of production together as a measure of total production in the economy. Economic output will always be
greater than GDP (also termed as value-added).

81
methods both attempt to assess the gross level of activity or expenditure in any economy.
They are not net measures that weigh benefits against costs, but are still useful to
understanding the size of businesses, projects, investments and economic sectors.

Definitions of direct, indirect and induced economic impact within the aviation sector can
be defined as follows:

Direct economic impact is employment, value-added or economic output that can be


attributed to the operation of air services to/from specific countries. In the aviation
sector, for example, the direct employment base includes employees of airlines, fixed
base operators, aircraft maintenance, and other firms located at airports.

Indirect economic impact is employment, value-added or economic output created


in industries that supply goods and services to air transport related firms. For example,
a wholesale food distribution company that supplies food to airlines would be part of
indirect economic impact.

Induced economic impact is employment, value-added or economic output


generated because of expenditures by individuals employed directly or indirectly by the
air transport industry. For example, if an air carrier employee decides to expand or re-
model his/her home, this would result in additional (induced) employment hours in the
general economy.

Total economic impact is the sum of direct, indirect and induced effects. The
multiplier (indirect and induced) economic impacts represent the maximum potential
stimulus to the economy resulting from air transport related activities.

Indirect and induced economic impacts are typically measured by the use of economic
multipliers which are derived from economic/statistical/accounting models of the general
economy. They come in a variety of forms and differ greatly in definition and application.
Thus, great care must be exercised in choosing the appropriate set of multipliers to use.

The use of multiplier analysis is limited by a number of factors, these are:

the accuracy of the structure and parameters of the underlying model

the level of unemployment in the economy

the assumption of constant returns to scale in production

the assumption that the economy's structure is static over time and

the assumption that there are no displacement effects.

Multiplier impacts must be interpreted with caution since they may be illusory when the
economy experiences high employment and output near industry capacity.

82
3. AVIATION SECTOR IMPACTS - DEFINITIONS
In order to quantify the impact of change in air freight and passenger traffic on the
aggregate performance of the economy, a set of coefficients was developed to estimate
the direct, indirect and induced economic impacts.

Aviation Sector World Regions

The economic impact of aviation can be different in different types of economies and in
different regions. Accordingly, this study identified 14 categories of nations based on a
combination of geographic location and country classifications used by international
organizations such as the United Nations, the OECD and the World Bank. Figure 1 provides
the fourteen world regions for the aviation sector economic impacts.

Figure 1: Aviation Sector Economic Impact World Regions

Developed Countries North America India Sub-Continent

Developed Countries Europe Developing Countries Mexico & Caribbean

Developed Countries Asia-Pacific Developing Countries Markets Latin


America
Emerging European Markets
Developing Countries Middle East
Emerging Markets Latin America
Developing Countries Africa
Emerging Markets Asia Pacific
Developing Countries Asia Pacific
China
Least Developed Countries

3. AVIATION SECTOR IMPACTS METHODOLOGY


The aviation sector ratios and economic impact multipliers were estimated from several
industry statistical publications and reports, individual airport economic impact studies61
and government data.

Existing industry data and reports which provided regional or global impacts included:

The Air Transport Action Group The Economic & Social Benefits of Air Transport (2004
data)

61
An advantage of individual airport economic impact studies is that the researcher typically has access to the most
detailed local data available and develops the most appropriate data and multipliers.

83
Airports Council International Europe The Social and Economic Impact of Airports in
Europe (2003 data)

Airports Council International North America The Economic Impact of U.S. Airports
(2002 data)

Airports Council International North America The Economic Impact of Canadian


Airports (2002 data)

Airports Council International 2005 Economic Survey

International Civil Aviation Organization Airports: Vital Catalyst for Economic Growth
(2003 data)

International Civil Aviation Organization Economic Contribution of Civil Aviation:


Ripples of Prosperity (1998 data)

Wilbur Smith Associates The Economic Impact of Civil Aviation on the U.S. Economy
(2000 data)

The ATAG study was used as the starting point for establishing indirect and induced
employment multipliers, as well as direct, indirect and induced GDP multipliers. The
numbers were generally consistent with other existing studies, as well as government
input-output tables and other published data sources. However, the ATAG study only
provided impacts for six world regions (North America, Europe, Latin America, Asia-
Pacific, Middle East and Africa). In order to provide a greater level of geographic
distinction within individual world regions, input-output data, employment and GDP
data for the transport industry was utilized. This data was generally available only at
the total transportation industry level, although some jurisdictions had detailed
aviation data available. This allows the model to provide separate economic impacts,
for example, for China or the India-sub continent, rather than using a broad set of
multipliers for the entire Asia-Pacific region.

Economic impact studies commissioned by individual airports were also analyzed to


provide additional detail and serve as a cross-check for the regional and global
studies. Airport economic impact studies were most readily available for airports in
North America and Europe.

In order to link changes in air passenger volumes to economic impact, a ratio of direct
employment to air passenger volumes was developed. The ratios were based primarily on
the ACI 2005 Economic Survey and ACI 2005 preliminary global traffic results. However,
because the ACI study included breakdowns for only five world regions (North America,
Europe, Asia/Pacific, Latin America/Caribbean and Africa/Middle East) and only limited
country level data was available for other world regions, the ratios are generally similar
within each geographic area.

84
The employment to passenger volume ratio has been converted to a workload unit basis in
order to be applicable to both air passenger and air cargo volumes. One workload unit is
the equivalent of one air passenger or 100 kilograms of air cargo, which loosely
corresponds to the weight of an average air passenger plus their baggage.

4. TOURISM SECTOR IMPACTS


Air service facilitates the arrival of both leisure and business tourists to a region or
country. The spending these tourists make in the destination country can create direct,
indirect and induced economic impacts in a wide range of tourism related industries:
hotels, restaurants, theatres, ground transportation, attractions, shopping, etc.

Countries were divided into tourism world regions according to geographic location and
development of the local tourism industry. A total of 13 tourism economic impact
categories were created as illustrated in Figure 2.

Figure 2: Tourism Sector Economic Impact World Regions

North America Well Developed China

Europe Well Developed India

Latin America Well Developed Europe Less Developed

Africa Well Developed Latin America Less Developed

Asia Pacific Well Developed Africa Less Developed

Mexico & Caribbean Asia Pacific Less Developed

Middle East

Tourism related expenditures, employment, GDP and multipliers were based primarily on
data published by major tourism organizations:

U.N. World Tourism Organization (UN-WTO) Compendium of Tourism Statistics (1999-


2003 data)

World Travel & Tourism Council (WTTC) Country League Tables (2005 data)

Additional individual country level data was obtained directly from national tourist
departments, statistical offices and academic papers.

In order to determine the economic impact of international tourists arriving at individual


countries by air transportation, various tourism ratios were developed including:

85
Inbound tourist arrivals as a percentage of total arrivals inbound tourist ratios were
based on total inbound/outbound tourist volumes reported by the UN-WTO. The ratios
are based on total international visitors traveling by all modes because air-only data
was not available. The percentages were estimated by region-pair. By applying the
appropriate ratio to the total increase in air traffic between two countries or regions, an
estimate of total inbound tourists can be calculated.

Average expenditure per international tourist visit international tourist expenditure


data was sourced from a combination of UN-WTO and WTTC publications. The data
includes all expenditures made by tourists within a destination country or region
including hotels, restaurants, sightseeing, local transportation, retail purchases, etc.,
but does not include purchases made in their home country prior to departure (e.g., air
transportation, package tours, etc.). The expenditure data was based on all
international visitors, including same-day visitors and visitors arriving by all modes.

Employment per $1 million of tourist expenditure total tourism related employment


was generally sourced from national tourism satellite accounts published by individual
countries. Because the employment figures were only available at the industry level
and not attributable to domestic versus international sectors, the employment ratios
are based on combined domestic and international data. The tourism data has been
adjusted to remove the air transport related employment in order to avoid double
counting the employment impacts already included in the air transport economic
impact above. Based on select country tourism satellite accounts which provided
employment by sector, an estimated 8 percent of employment was removed to account
for air transport related jobs.

In order to establish the total economic impacts on the broader economy, multipliers were
developed from WTTC data sources and tourism economic impact studies for individual
tourism markets.

5. CATALYTIC IMPACTS
Several studies have examined the impact of air transportation on business location,
economic development, investment and employment. Some have examined the
importance of air transportation in the location decision making of businesses. For
example, the Atlanta Chamber of Commerce (1988) surveyed 264 foreign-based firms
located in and around Atlanta and found that the extent of direct international air service
was the third most important factor in location decisions. Ernst and Young, reviewing the
decisions of 57 European companies, found that the extent of air transport network was
the third most important factor in the selection process. The Amsterdam Chamber of
Commerce found that the availability of an airport was one of five key factors considered
in company relocation decisions.

The studies described above demonstrate that air transportation plays an important role in
business location decisions. Other studies have uncovered empirical evidence

86
demonstrating a strong linkage between air service and economic development and
investment.

Irwin and Kasarda (1991) examined the relationship between the structure of airline
networks and employment growth at 104 metropolitan areas in the United States. Using
data for a 30 year period, the researchers conducted regression analysis relating
employment in the manufacturing and service sectors in a metropolitan area to number of
explanatory factors including population, road infrastructure, telecommunication
infrastructure and measures of airline network serving the area.62 This regression analysis
showed that expansion of the airline network serving a region had a significant positive
impact on employment in that region. The effect was particularly significant in the service
sector. Furthermore, analysis was conducted (using nonrecursive models) which confirmed
that increases in the airline network were a cause rather than a consequence of this
employment growth.

Hansen and Gerstein (1991) investigated the relationship between Japanese air service to
the United States and Japanese direct investment in the United States. Using data from
1982 to 1987, the analysis related the amount of Japanese investment in each U.S. state
to measures of level of air service operated between Japan and that state (and other
background factors). The analysis found a significant positive relationship between
investment and air service. The results also suggested that the amount of service provided
by Japanese carriers had a larger impact on investment than service provided by U.S.
carriers. The issue of causality is also addressed (i.e., does more air service lead to
greater investment or does greater investment lead to more air service), with the authors
concluding that the evidence indicates that air service impacts on investment rather than
the other way around. The authors concluded that better air service supports the input
needs (i.e., labor and materials) of the Japanese ventures in the U.S. and enables greater
awareness and information flows in Japan for U.S. regions.

Button, Lall, Stough and Trice (1999) examined empirically the link between high-tech
employment in a region and whether the region is served by a hub airport. Using data
from 321 U.S. metropolitan areas in 1994, the authors regressed high-tech employment
against a number of controlling factors, including a dummy indicating that the region was
served by a hub airport.63 The analysis found that the presence of a hub airport increased
high tech employment by an average of 12,000 jobs in a region. The authors also
addressed the issue of causality (i.e., does the presence of a hub airport lead to more
employment, or does higher employment in a region increase the likelihood of a hub
airport being developed). Using the Granger causality test, the authors found that there

62
The researchers used specific definitions of manufacturing and service sector employment which excluded most of
the employment associated with the aviation sector and tourism.
63
High-tech employment included IT, telecoms, biotechnology, electronics, and certain types of high-value
manufacturing. It excludes aviation (except for manufacturing) and tourism. The study used the Federal Aviation
Authoritys standard definition for a hub airport (using this definition, there were 56 hubs in the U.S. in 1994).

87
was statistically significant evidence that the presence of a hub airport caused an increase
in high-tech employment.

A similar study by Button and Taylor (2000) examined the link between international air
service and economic development. Using data for 41 metropolitan areas in the U.S., the
authors regressed new economy employment against a number of control factors,
including the number of direct routes to Europe offered by airports in the region.64 The
analysis found that there was a strong and significant relationship between employment
and air services to Europe. The impact was largest for regions which initially had very
limited services to Europe. For example, increasing the number of European routes served
from 3 to 4 (40,000 additional passengers per annum) generated approximately 2,900
new economy jobs. However, increasing the number of routes served from 20 to 21
(10,000 passengers) generated 440 new economy jobs. This analysis suggests that each
1,000 enplaning/deplaning (E/D) passengers increases employment by 44 to 73 jobs,
depending on the number of routes already served. To address the issue of causality (and
to allow for a lagged response to the new air service), employment in 1996 was regressed
against the number of routes in 1994.

Another study, Brueckner (2002), also looked impact of air service on employment in the
U.S. The author regressed employment in 94 metropolitan areas in the U.S. against a
number of factors, including measures of air service. The analysis found that a 10 percent
increase in passenger enplanements in a metropolitan area leads to an approximately 1
percent increase in employment in service-related industries. However, the analysis found
that there was no impact on manufacturing and other goods-related employment,
suggesting that air travel is less important to these industries than it is to service-related
industries. Using Chicago as an example, the author estimates that each additional 1,000
E/D passengers increases employment by 22 jobs.65 The author also conducted analysis
(Granger causality test) to address the issue of direction of causality and found statistically
significant evidence that air traffic was a cause of employment.

A study commissioned by EUROCONTROL (2005) examined the catalytic effects of air


transportation in Europe.66 The study examined the contribution of air transportation to
tourism, trade, location/investment decisions and productivity. It estimated the net
contribution of air transportation to trade (i.e., export minus imports) to be 55.7 Billion in
2003 across the 25 current EU members, or approximate 0.6 percent of GDP. The study

64
Similar to high-tech employment, new economy employment included IT, telecoms, biotechnology, electronics,
and certain types of high-value manufacturing, but generally excluded aviation (except for manufacturing) and
tourism.
65
Unlike some of the previous studies, the definition of employment does include aviation sector and tourism
employment.
66
EUROCONTROL is a civil and military organization established in 1963 to facilitate a safe, seamless pan-European
Air Traffic Management (ATM) system. While the initial focus of the organization was on safety and operations, its
remit has expanded over time to include capacity management and development, operating costs, and fees and
charges. EUROCONTROL is not an EU institution, but includes nearly all the EU members, as well as countries
outside of the EU such as Switzerland, Turkey and Norway.

88
analyzed the relationship between air transportation and business investment. Using
Equilibrium Correction Mechanism (ECM) equation analysis of data from EU member
countries, and controlling for other factors, the results indicated that a 10 percent increase
in air transportation usage (relative to GDP) will tend to increase business investment by
1.6 percent in the long run (the impact takes approximately five years to fully manifest).
The authors estimated that between 1994 and 2003, air transportation increased business
investment by 5.8 percent in the 25 EU member countries, worth 66 billion.

The study also investigated the relationship between air transportation and productivity.
Again using ECM analysis of data from EU member countries, the analysis indicated that a
10 percent increase in air transport usage (relative to GDP) increases underlying
productivity by 0.56 percent in the long run. Productivity is boosted by firms being better
able to exploit economies of scale, access a wider pool of labor, and experience greater
exposure to competition. This promotes innovation and efficiency.

Combining these impacts together (excluding the impact of tourism), the study estimates
that air transportation contributed to a 4.6 percent growth in GDP across the 25 EU
members between 1994 and 2003, worth a total of 480 billion.

The studies described above provide strong evidence that air transportation is a significant
contributor to economic growth and development. They have also been used in this study
as the basis for estimating the catalytic impact of air transportation.

The study by Button and Taylor (2000) found that 44-73 jobs are generated for each
additional 1,000 E/D passengers while the study by Brueckner (2002) found that 22 jobs
were generated by each additional 1,000 E/D passengers.

On the basis of these studies, it has been assumed that approximately 30 full-
time equivalent jobs are generated for each additional 1,000 enplaning/deplaning
passenger.

G. SUMMARY OF MODEL

The passenger, freight and economic impact models, while seemingly separate entities,
are components of a larger and integrated framework. This section summarizes the
interactions between each part, and how they together form a model of liberalization with
worldwide applicability relationships binding these parts, and the framework culminates in
a model of liberalization with worldwide applicability.

89
Figure 3 shows a simplified schematic of the adopted approach. The algorithm includes
separate single-equation models for passenger traffic and air freight67.

Any
Figure 3: Schematic of Adopted Approach Country
Pair

Airline Pax. by Bilateral Socio- Air Freight Current Aircraft Economic


Schedules Country- Agreements economic by Country- Bilateral Capacity, Impact
Pair Variables Pair Distance Coefficients

Passenger Air Freight Liberalization


Model Model Scenario

Forecast Percentage Air Freight


Passengers by Traffic Increase Forecast
Country-Pair

NO Do We Know
Current Traffic?

YES

Passenger Freight
Increase Directional
Imbalances

Number of
New Flights

Validation New Air Freight Capacity

Air Freight Traffic Increase

Pax. Traffic Tourism Catalytic Air Freight


Employment Employment Employment Employment

Pax. Traffic Tourism Catalytic Air Freight


GDP GDP GDP GDP

67
The passenger and air freight traffic demands between any country-pair depend on a wide range of factors, each
of which belongs to either of the broad categories of supply or demand. According to economic theory, the
price variable equilibrates both factors. This means that a particular set of price-quantity combinations in isolation
says nothing about either supply or demand. Unless the researcher exercises considerable caution, and uses
specialized techniques, efforts to obtain information about the underlying relationships could generate misleading
and inaccurate results. The model does not use price to explain demand, partly because suitable information on
prices and costs was not available. Rather, it applied a reduced form approach in which traffic flows were deemed
as determined solely by factors outside the model.

90
The passenger model relates traffic to the scale of economic activity and the constraints
posed by air service agreements. The model was developed from information about 800
country-pairs. The model can quantify the total number of passengers traveling between
literally any country-pair, including those for which no traffic data is available. Since the
model includes information about the bilateral agreements, it can readily estimate the
increase in traffic resulting from liberalization.

Passenger traffic for some country-pairs is already available. That part of the model that
includes terms for the bilateral can produce an estimate of the percentage increase in
traffic corresponding to different degrees and types of liberalization. This percentage can
be applied to the current traffic base. If the base traffic is not available, the model can
generate forecasts of the traffic before and after the liberalization. Either approach will
calculate the number of new passengers resulting from the agreement. The resulting
economic benefits are calculated from either measure.

The incremental passenger traffic is subject to a series of tests of reasonableness. If the


country-pair in question already obtains direct service, the model assumes that seat
capacity will expand in proportion to the new traffic. If not, the algorithm examines
prevailing relationships between aircraft size and useful range, and calculates the number
of flights required to meet a seat and load factor target. If demands are too small to
justify a certain minimal level of service, the model concludes that no traffic growth can
result, and that the liberalization will generate no immediate economic benefits.

The complexities of air freight limit the methodologies used and the applicability of the
final results. The algorithm therefore pursues the less ambitious goal of modeling the
impacts of increased air freight that result from an expansion of passenger services
following liberalization.

Two conditions may occur. The country-pair may already have scheduled passenger
services. Since the model assumes no change in the mix of aircraft, air freight capacity
will rise in the same proportion. Airlines and freight forwarders will sell this additional
capacity, and air freight traffic will grow accordingly. The similar passenger and passenger
growth rates in no respect indicate that the demands grow in the same manner, but,
rather, that belly air freight volumes are intimately linked to passenger marketing
decisions, and is sold as a form of incidental revenue. A second case governs
country-pairs with no current passenger flights. The growth of passenger traffic resulting
from the liberalization could lead to new services. The frequencies would be determined
by passenger demands; the aircraft type, with its pre-defined air freight capacity, would
be assigned by the passenger model. Under the by-product assumption, air freight flows
would passively adjust to use the additional capacity. The capacity in the heavy
direction would be fully utilized; incremental freight traffic in the light direction would
reflect the directional imbalances estimated by the econometric model.

91
Matrices developed from several sources define the relationships between additional
passenger and air freight traffic, and increases in employment and Gross Domestic
Product. They consider effects that arise from increased passenger travel, increased
international tourism, a growth of air freight, and catalytic effects. The coefficients apply
to each nation in a country-pair.

H. THE MODEL - ISSUES

The cross-sectional modeling approach demonstrates that economic fundamentals such as


Gross Domestic Product, the level of trade in services, and geographical variables are by
far the most important determinants of traffic between any country-pair.

Variables that reflect the severity of the artificial restrictions posed by bilateral agreements
play a smaller role in defining traffic. Their influences are unquestionably significant, both
in the narrow statistical definition and in the broader social sense. They are the only
variables over which our societies can exercise a large degree of control, and change
quickly at relatively low cost.

The research included developing a simple and highly structured measure of the
liberalness-restrictiveness of any particular bilateral. It applied a composite score of each
attribute number of airlines designated, pricing provisions, capacity clauses, route
schedules, fifth freedom rights, etc., into a penalty index ranging continuously between
zero and unity. The total traffic was expressed as a product of this function, and as a
function of the purely socioeconomic variables. This specification necessitated a non-linear
estimation process, which converged rapidly and displayed the necessary mathematical
conditions.68 Several other linear and non-linear methods isolated the regulatory/bilateral
influences.

No benefit was gained from the additional complexity of these approaches, either in terms
of desirable statistical properties or in usefulness of the results. The model stated above,
estimated with a generalized least squares algorithm, offered the best tradeoffs between
simplicity, statistical goodness of fit, and an ability to make useful inferences about
liberalization policy.

The air freight model served a limited role. A view of air freight as strictly subordinated to
passenger traffic, with capacity growth tied to additional passenger volumes, cannot, by
definition, simulate air freight as an independent business. Air freight traffic patterns are
very specific to the situation. They are governed by the habits of individual shippers,
airlines, exchange rates, the demands for specific commodities, consolidator business
practices and very unique local circumstances. For example, IATA reports that Japan

68
A vector of zero first derivatives, stability, and a negative-definite Hessian.

92
exchanges 25 percent more air freight with the Marshall Islands than with the United
Kingdom.69 Very strong and situation-specific conditions govern many air freight markets.

The models yielded statistically satisfactory results. The decisive majority of the
coefficients were significant. The signs were as expected; for example, the
positive coefficient of the GDP variable shows that large traffic volumes are
associated with high levels of economic activity. Restrictive bilateral agreements
do constrain traffic.

The r-square measures were weak. The r-square, which ranges from zero to
unity, measures the proportion of the total variation of series of interest
(passenger or directional air freight by country-pair) that is explained by the
model. The low values obtained show that other factors besides GDP, bilateral
agreements and other variables explain traffic.

Cross-sectional models of the type developed often have low r-squareds. The
models developed in this Study are not atypical. In contrast, a time series
model, that evaluates a single process over specific period, can easily develop an
r-square exceeding .9. This result, however attractive, in no way credits the
underlying research.

The estimation process considered many other variables besides those finally
selected. While many different specifications are reasonable, none resulted in
such a large improvement in the goodness of fit to generate a cosmetically
attractive r-square. The biggest gap in the model likely concerns issues of
personal taste. The Nouveaux Beaujolais wine, which is shipped by air in large
quantities, is very popular around the world, and part of an annual ritual that
just is. Central America is popular for ecotourists, while Walt Disney World has
a unique international identification. It appears that the demand for air travel
and air freight depends on a few very important and easily definable quantities,
but also on a vast number of very country-specific preferences and customs.
While individually small, they are collectively important. They are difficult to
model effectively.

Any effort to raise the r-square through a diligent search for something that
works is poor methodology. The items in question will inevitably be found, but
the underlying purpose of the search, and the inevitability of the outcome, make
the results trivial, if not deliberately misleading.

69
Source: International Air Transport Association, Freight Forecast 2005-2009, (Geneva, 2005) Page 151.

93
The research shows that good statistical fit, as indicated by an r-square measure
close to unity, is very unlikely. It is in the very nature of commercial aviation; in
its financial volatility, the huge variation of economic and social factors
throughout the world, and the changing industry structure, that is very complex,
and defies any simple modeling or any easy predictability. There was no initial
reason to expect it to conform closely to any formal model.

The air freight model did not effectively accommodate the problem of directional
imbalances. If traffic volumes and directional ratios were provided by the data, the model
assumed that post-liberalization directional patterns would continue. If country-pair data is
not available, the model must rely on the regression model to impute directional ratios.
The coefficients for the exporting and importing nations are too similar to generate the
types of imbalances so prevalent on major trade corridors such as the North Pacific.

The sheer complexity of air freight poses a challenge to any effort to develop a simple,
compact model that can accurately depict the consequences of liberalization in any market
at all. Such an algorithm must consider the minutiae of the importing and exporting
economies, alternate country markets and suppliers, directional imbalances, the
availability of belly capacity on passenger aircraft, exchange rates, intermodal services,
the consolidation practices of forwarders, and other details. It might even need to address
the practices of specific importers and exporters. Air freight services often depend on fifth
and sixth freedom markets, or international feed by trucks, so no country-pair or direction
should be viewed in isolation. A bottom-up approach for air freight, developing forecasts
for a single market from a micro-analysis of many variables, both present and historical, is
preferable to a top-down method that seeks worldwide patterns of behavior.

Other studies and models have examined the consequences of liberalization for a small,
localized and clearly defined market. These models can therefore incorporate many details
about current traffic volumes, socioeconomic variables, the current level of air service and
the applicable bilateral agreements. Their findings reflect a micro-analysis of the different
minutiae governing carrier behavior. They are data-intensive.

In contrast, the approach outlined above is data-extensive. Since it must apply for any
arbitrary country-pair, there is no opportunity to assemble detailed economic data and
calibrate a model for a specific market. Rather, it uses a relatively small amount of data
for a particular country-pair. However, it considers over a thousand country-pairs,
spanning a huge range of conditions. The very diversity of the sample supports a model of
extremely wide applicability. However, this versatility comes with a penalty. The
predictions will be less accurate, and less focused to the unique circumstances of a
market, than a traditional route-specific before liberalization-after liberalization analysis.

94
The Economic Impact of
Air Service Liberalization

V. Findings and
Conclusions
V. FINDINGS AND CONCLUSIONS

The principal findings are outlined in the Executive Summary Section of this study. There
is little doubt that liberalization of aviation bilateral air services agreements, and the
deregulation of domestic aviation markets, has brought substantial benefits to economies
and consumers. For these reasons, the vast majority of world governments have given
either explicit or implicit support for the concept, if not always the practice of liberalizing
air service.

Our research, and the design and development of the economic model (reviewed in
Section IV), lead to the conclusions that:

1. Liberalizing bilateral air service agreements (ASAs) can generate significant gains,
worldwide, in terms of expanding economies and creating employment.

2. Even gradual expansion of rights under ASAs can lead to significant gains for
carriers, consumers, and the communities at large, e.g. the recent U.S.-China
agreement.

3. There is a continuing opportunity for governments of the world to liberalize air


services, and thereby expand their economies and create jobs for their
constituents. Our model simulation of the scope of the change that would take
place if 320 ASAs were fully liberalized produced economic benefits almost as large
as the GDP of Brazil.

4. A U.S.-European Union First Phase Air Transport Agreement would free an


additional four major country-pair markets, i.e. the U.S. to the U.K., Spain,
Ireland, and Greece. Such an agreement would generate an additional 117,000
jobs and $7.8 billion between the U.S. and the U.K. alone.

5. Market fragmentation is increasingly offering non-stop opportunities to consumers


between countries and cities where traffic would have been insufficient 10 years
ago. Moreover, longer-range aircraft are driving this fragmentation to connect
points almost halfway around the world.

6. The creation of the Single Aviation Market in Europe in 1993 has been one of the
single most prominent success stories in aviation deregulation. Traffic post 1994
ultimately grew at average rates that were double those of pre-1994 years.

In light of the above, it is evident that the continued education of the world community on
the benefits of free trade in aviation is a worthwhile undertaking.

95
The Economic Impact of
Air Service Liberalization

Appendix
APPENDIX A: THE INTERNATIONAL AVIATION FRAMEWORK

A. INTRODUCTION

This study has examined the effects of political and institutional restrictions on
international air service. While these constraints vary widely by nation, market and city-
pair, most originate from the bilateral air service agreements and the associated
regulatory machinery that governs international air commerce. This Appendix describes
the political and regulatory institutions of international aviation; their history, structure,
and ways in which they can limit the free flow of people and goods. The fundamental issue
in this study is that commercial aviation is one of the very few sectors that has been
exempted from the principles of economic liberalism, free trade and atomistic decision
making by the market.

B. FREE TRADE, PERSONAL FREEDOM AND COMMERCIAL AVIATION

Personal freedom and economic efficiency require a free flow of resources (people, capital,
land, commodities, etc.) between alternate uses. Most nations have therefore adopted
liberal economic policies that allow and encourage individuals, and groups of individuals,
to develop their opportunities to the fullest. While tariff and non-tariff barriers are
widespread, most countries extend these freedoms to international trade, where any
activity is permitted unless specifically forbidden.

While embracing the ideal of economic freedom, all societies have accepted in principle
that certain constraints on individual liberties are reasonable and necessary. Such limits
may curb abuses of power (e.g. the legal system and the laws that regulate common
behavior), eliminate socially destructive activities (trading in recreational drugs or
explosives) or promote collective security (military conscription). Others assist members
of the society judged unable to make sound decisions (compulsory school attendance for
minors), protect the weak (rules reserving certain seats on rapid transit systems for sick
or elderly persons) or direct resources to activities that could be overlooked by individual
decision makers (assessing taxes to support public programs). Sometimes, individual
liberties are curbed in order to raise or stabilize incomes of particular groups (rules
restricting free trade in agricultural products). While such rules are universal, they are
purely exceptions to a wide degree of freedom available to the inhabitants of virtually all
nations, including those under even the most rigid centralized planning.

Aviation is an exception. Political and legal interference in the free flow of civil aviation
opportunities has often been embraced in many parts of the world. Commercial airlines
have historically been viewed by governments as playing an important role in national
defense and political sovereignty. Their special strategic role ostensibly made them too

A-1
important to be governed by market forces. Furthermore, any industry that so readily
slipped the surly bonds of earth could pose clear safety issues, which might be
undermined if operators were exposed to market pressures. These concerns serve as the
basis for a worldwide legal framework that is still quite intrusive.

Even well before the Second World War, commercial aviation had proven its potential. It
was used for premium air mail. Several nations used air transport to integrate their far-
flung colonial holdings. Aviation had proven crucial in both world wars. The lack of any
worldwide regulatory framework posed a challenge to all nations. In 1944, delegates met
in Chicago to establish a mechanism for further expansion.

C. THE CHICAGO CONVENTION AND BILATERAL AIR SERVICE AGREEMENTS

The Chicago Convention of 1944 laid down the current framework for international
aviation. The Convention saw a very close association between national governments and
the airlines that would operate the international services. In the immediate postwar era,
most airlines were owned by their respective governments, and were considered to play
important roles in national defense, showing the flag by operating high profile but
money-losing routes, and serving as important tools of foreign policy. The very high fares
made them largely unimportant as a mode of transportation. Their roles of generating
income as businesses and fostering economic development in the communities they
served also were largely irrelevant.

The Convention stipulated that two nations, seeking to be linked by commercial services,
would negotiate the terms through concluding a bilateral air service agreement or
bilateral. This would specify the conditions under which the proposed services would
operate, in terms of the privileges granted by either signatory country to the airline or
airlines of the other party. The agreement would cover such items as:

The allowable routes that could be operated. This could range from a general statement
such as any point in Country A to any point in Country B to an exhaustively detailed
specification of individual airports, and what points could or could not be combined on a
particular flight and in what order.

Whether the flights between countries A and B could involve third countries. The
agreement would often state that cities in third countries could be served only as
enroute stops (Intermediates) on flights linking nations A and B, or by flights that
continue onward after operating the A-B or B-A sectors (Beyonds).70 In many
instances, the bilateral agreements specify in meticulous detail what services could be
operated, whether the airline(s) can carry revenue traffic on such flights, what

70
An airline of country A seeking to operate a flight from A to B and on to nation C would require that both the A-B
and A-C bilateral agreements would permit the service.

A-2
combinations of intermediate/beyond points are permissible, frequency restrictions and
other minutiae.

The number of airlines that each country may designate to the international services.

The capacity that the airlines of each country could offer.

A method for setting fares on the route. The agreement would specify the conditions
necessary for a fare proposed by the airline of one country to become operative.

Various doing business issues such as repatriation of currencies, the ability to select
handling agents at foreign airports and use of computer reservations systems.

Once the two nations have concluded the bilateral agreement, each country is permitted
to select the airline or airlines to operate the routes. This capability is considered an
important expression of national sovereignty. Some agreements do restrict these choices.
Virtually all bilateral agreements allow a nation to designate only those airlines that can
demonstrate that they are corporate citizens of that country. This requires that they be
substantially owned and under effective control by nationals of that country. This rule
reflects the traditional identification of airlines with specific countries; a situation prevailing
when the Convention was signed. It does not accommodate a regime where airlines are an
industry like any other, with multinational ownership, control and operations.

The Chicago Convention framework clearly distinguishes between international and


domestic services. Domestic services are strictly a matter for the respective national
government. Virtually all countries prohibit foreign airlines from offering purely domestic
services. They may, however, allow foreign operators to operate international flights to
third countries, such as Singapore Airlines Los Angeles-Tokyo-Singapore route.

The Chicago Convention and the underlying bilateral agreements have become
increasingly controversial. The next section summarizes how this machinery can limit
international aviation.

D. BILATERAL AGREEMENTS - ISSUES

The framework of the Chicago Convention has proven to be very durable and flexible. It
can allow a wide range of market regimes, from highly restrictive agreements with rigidly
defined descriptions of allowable city-pairs, a precise definition of allowable capacities and
frequencies and intrusive pricing methods; to the very liberal open skies agreements
that allow free entry of airlines of either signatory nation to any route, unrestricted
capacity, and full pricing freedom.

The system of bilateral agreements poses a growing obstacle to the full development of
worldwide air commerce. The shortcomings originate from three sources:

A-3
1. THE CONTENT OF THE AGREEMENTS
Most bilateral agreements constrain the routes that airlines may fly, the number of airlines
that could compete, their capacities and the rates they may charge. The agreements may
rule out some otherwise viable services, perpetuate certain inefficient services, and
deprive the airlines of the flexibility needed to manage. Sometimes, the relevant airlines
may fully recognize the need for changes to the agreement, but the timing will be subject
to the willingness of governments to make changes.

2. THE ASSUMPTIONS UNDERLYING THE FRAMEWORK OF AGREEMENTS


The machinery laid out in the Chicago Convention assumes a close identification between
an airline and a nation state. Every airline operating international services is assumed to
be closely associated with one nation through ownership and control by nationals. This
perpetuates a fragmented and often inefficient industry structure, because an airline
seeking foreign capital or a foreign merger partner could lose the ability to operate
international flights. A country usually cannot designate a foreign country to exercise the
rights it has obtained through bilateral negotiations. While commercial aviation is arguably
among the most cosmopolitan and outward-looking industries in the world, its entities
cannot evolve into genuine multinationals.

Many leading international airlines have joined strategic alliances such as the Star Alliance,
SkyTeam and oneworld. These associations include airlines from throughout the world,
and allow their members to operate a largely seamless, self-feeding global network. They
increasingly coordinate their products on a worldwide basis, involving dozens of individual
countries. The Chicago Convention framework, which operates on a country-pair basis, is
increasingly unable to accommodate the multi-airline alliances.

3. THE IMPACT OF THE BILATERAL FRAMEWORK AIRLINE-GOVERNMENT RELATIONS AND AIRLINE BEHAVIOR
This issue is arguably the most complicated aspect of the current framework. Restrictive
bilateral agreements must be viewed not merely as static barriers of trade precluding
certain flights, and imposing a degree of friction on market growth, but as active features
of the market landscape. They have a dynamic impact on the market, creating a
partnership between governments and airlines, and affecting the competitive advantages
of different airlines. Airlines often behave differently in a restricted market, creating a host
of distortions.

The system of bilateral agreements has given for-profit airlines a form of economic
protection backed by the full coercive powers of the state. It often allows, even
encourages, airlines to block specific competitors or services; not to the benefit of the
economy, the national interest, or personal safety, but for airline managers, shareholders
and employees. It can foster an unwholesome relationship between governments and the
airlines they are protecting.

A-4
This rationale was valid for the years immediately following World War II. Technological
advances have dramatically lowered the cost of air travel, and it has since expanded far
beyond the limited role foreseen in the Convention. The widespread privatization of
airlines, while far from complete, further weakens the close government-carrier
relationships that governed the industry in the 40s and 50s. Commercial aviation has
become a for-profit industry like any other, and no longer requires a pervasive system of
economic regulation. Most importantly, the emergence of mass travel, and the widespread
use of air freight for high priority shipments, has created a huge group of new
stakeholders: the users. With users of different communities either well or poorly served
by airlines, bilateral agreements have become an important factor in regional economic
development, and a source of dissension between the have- and the have-not
communities.

Despite these changes, the current framework still requires a close association between
airlines and governments. An airline wishing to begin an international service must obtain
the permission of the national government. If no bilateral agreement exists, that
government must approach the government of the other nation to request bilateral
negotiations. That nation in turn will seek input from its own airlines as to the ideal terms
of the agreement. If its airlines have no interest in obtaining parallel rights, or have no
wish list of their own, the effort could easily falter at this stage. In some cases, it will
simply be unable to obtain the rights it seeks. In others, an airline will not propose to
serve a desired route, since it may view the pursuit of such rights as inevitably futile.

The horse-trading of bilateral negotiations, as envisioned by the Convention, implicitly


defines the balance of benefits in a very narrow fashion. There is a tendency to
consider only the airlines as stakeholders and to neglect the broader needs of
passengers, shippers, and others who depend directly or indirectly on high quality
transportation. Government air service policy officers are fully aware of the impact of
air service on regional development. However, it is very difficult to incorporate an
economic development role into any bilateral negotiating position, or to define what
negotiators actually seek. Regional economic concerns are often overlooked in this
vacuum. The interests of airlines, and the benefits and costs of different tradeoffs, can
usually be estimated with considerable accuracy. The negotiating mandate is usually
some variation of get as much as possible for our side, and give as little as necessary
to their side.

Airlines can easily estimate the impact of any change in an air service agreement. A
drop in profits that would result from allowing a foreign carrier unreciprocated entry to
a city may be far smaller than the economic benefits the new flight would bring.
However, they would be borne entirely by a strong and sophisticated entity that could
raise a powerful resistance. The benefits, while much larger, are diffused across many
individuals. On a per-person basis, they would be too small to encourage any strong
support for the change.

A-5
Good air negotiators are shrewd, tough and highly competitive people. They often
judge themselves by their ability to make as few concessions as possible. This
sometimes prompts them to restrict foreign carrier operations in their own countries to
the fullest extent possible. This means that they will only permit new services by
foreign airlines if it is a necessary step for obtaining rights for their own carriers.

The commercial sensitivity of bilateral agreements and the manner in which they help
or hinder communities obtain air service makes them potentially very controversial.
Some U.S. negotiations are accompanied by intense lobbying, placing the persons and
institutions involved in international aviation in an unwanted limelight. As a result,
many countries do not publicize even important information about certain bilateral
agreements. The communities then do not even know the nature of the air service
constraints that they face.

Airlines often compete for the authorities granted by restrictive bilateral agreements.
Many airlines are closely identified with certain communities or interest groups. The
allocation processes can make political pressures almost inevitable.

In some circumstances, aviation negotiators will conclude a restrictive agreement, but


also incorporate a flexibility to accommodate market growth. For example an
agreement might allow only one airline from each nation to serve a particular route.
However, as traffic increases, negotiators may wish to permit greater competition. One
means is to allow further designations after the traffic attains a certain level. While well
meaning, such measures are often totally anti-competitive. The two airlines operating
the route are fully aware of the provision, and would not welcome further competition.
They have a strong incentive not to develop the route; to manage fares and capacity so
that traffic remains just below the threshold. As a duopoly, they can tacitly coordinate
their behavior to ensure that traffic never reaches the threshold. Since the bilateral is
already very restrictive and likely bars service for routes that could support it, such a
distortion on one of the few city-pairs with competitive flights greatly exacerbates the
adverse consequences of the regime.

The above discussion has centered on how restrictive bilateral agreements impede
market growth. However, a well functioning market by definition can redirect capital
from unproductive uses to superior ones. This means that it must also accommodate
shrinkage. If a scarce authority has a use it or lose it provision, an airline may be
reluctant to eliminate service even if a particular market is no longer profitable. It
would permanently lose the authority. It therefore has an incentive to continue serving
a money-losing market rather than reallocating resources to an unrestricted and
profitable use. The airline crisis that followed the terrorist attacks on New York and
Washington has highlighted the industrys financial vulnerability. Restrictive bilaterals
not only impede market entry; they also constrain market exit. They can exacerbate
the consequences of any market shocks.

A-6
E. CONCLUSIONS

The bilateral agreements that govern international air commerce are a growing
anachronism. The agreements are based on a small, fragile group of quasi-government
entities playing a diplomatic and military role, with negligible commercial significance. The
agreements, and the Chicago Convention framework are at variance with the development
of commercial aviation at large, where airlines must maximize operating efficiencies, and
shareholders demand an acceptable return on their investment. The bilateral agreements
are also inconsistent with much of the current thinking on the importance of free trade,
and the appropriate roles of government and for-profit private industry.

The bilateral agreements cannot be viewed only in a static sense, of allowing certain
flights but forbidding others. Rather, the institutional machinery distorts the relationships
between government regulators and airlines. It often results in very different patterns of
airline competition, and gives one airline a strong influence on the actions of another. If
the airline sector is inherently oligopolistic, such interdependencies have the capability of
distorting competition, even in the absence of pervasive regulation of international
markets.

The purpose of this Appendix has been to describe the system of bilateral agreements. As
demonstrated by the case histories, no generalizations are possible. The Chicago
Convention and the associated bilateral framework can support both very liberal and
totally oppressive agreements. In some cases, the agreements pose a severe impediment
to trade. In other markets, the bilateral agreements are virtually irrelevant; no airline
wishes to serve the particular country-pair. The relationships between the airlines and the
regulators vary between nations. They range from the airlines de facto control of the
negotiations, to a strict arms length and sometimes adversarial association. These
variations make the goal of this study, that of quantifying the economic value generated
by liberalizing bilateral agreements for any arbitrary country-pair, particularly ambitious.

A-7
APPENDIX B: GLOSSARY OF TERMS
Bermuda Agreement
In 1946, the United States and the United Kingdom negotiated one of the first air service
agreements under the Chicago Convention. The agreement, signed in Bermuda, included
capacity and pricing controls. According to the standards of 2006, it is a restrictive
structure. The so-called Bermuda I agreement has served as a prototype for many
subsequent agreements. In 1977, the Bermuda II Agreement, again involving the United
States and the United Kingdom, was similar to its predecessor in most respects, but
included restrictions of multiple designation, and provisions for capacity and all-cargo
services.

Bilateral Air Services Agreement


(See Appendix B, International Aviation Framework)

Catalytic Impacts
Catalytic impacts represent a fourth, and often difficult to measure, benefit to a particular
economic activity. The continued existence of the activity could cause long term changes
in the societys expectations. People observe the activity, assume its continued existence,
and modify their behavior accordingly. They then pursue new interests which would not be
possible in the absence of the original stimulus. For example, the presence of an airport
with commercial air services may make the community more attractive as a location for a
branch plant. Potential exporters could be offered low air freight rates to overseas
destinations, which would make them newly competitive. Neither the new businesses nor
the exporters need have any apparent relationship to commercial aviation, except as
customers.

Most traditional economic impact analyses allude to catalytic effects, but are unable to
quantify these benefits to any meaningful extent. Economic impact studies, which focus on
direct/indirect/induced benefits, follow a static and timeless perspective. Something,
usually the facility or service being studied, injects funds into an area from outside. The
region responds immediately with changes in expenditure, but the structure of
expectations and the economic base remain unchanged. The catalytic impacts represent a
dynamic adjustment, in which the underlying fabric of the region adjusts over time in
response to the activity.

Code Sharing
Code sharing allows two or more airlines to market their services as a single entity. One
airline can then sell tickets to points it does not actually serve, and contract with other
friendly carriers to operate routes on its behalf. For example, U.S. Airways has a code
sharing agreement with Chautauqua Airlines that allows it to sell T.F. Green-Elmira, N.Y.,
transportation on a single ticket and under a through fare. A passenger will fly by U.S.
Airways to Philadelphia, and connect to a Chautauqua Airlines flight to Elmira. The service
on the latter segment is branded to match the U.S. Airways product as closely as possible.

B-1
The passenger might travel to Washington Dulles on Air Wisconsin, thence to Copenhagen
on SAS, and continuing to Berlin on Lufthansa all under the UA code.

Computer Reservations Systems


Computer reservations systems (CRSs) are the intermediary between the consumer and
supplier, often operated by a travel agent to construct itineraries in response to passenger
requests. They follow a series of algorithms to rank the competing routings, price, etc.,
and display them to the prospective passenger. The algorithms place a strong weight on
connections that do not require an enroute change of airline. They weight the two-digit
codes for the airlines operating each segment to determine if the routing involves only one
airline. Where code sharing occurs, each flight bears several airline codes. For example,
the Washington-Copenhagen leg will bear the code of its own operator SAS (SK), United
(UA) and Lufthansa (LH).

Deregulation
The Airline Deregulation Act in the U.S. took effect in 1978. It abolished all governmental
controls on market entry and pricing. Under the previous regime, airlines were prohibited
from serving any market unless specifically allowed to do so by the Civil Aeronautics Board
(CAB), a federal agency. Prices were strictly controlled. No new airlines could be
established, unless they served purely intrastate routes (where federal jurisdiction did not
apply) or operated very small aircraft. Unprofitable carriers were strengthened by
favorable route awards. Federal regulation resulted in an industry where carriers competed
on frills (e.g. in-flight service, crew uniforms) and scheduling rather than on price.
Regulation spawned a paternalistic corporate culture and employee expectations for high
incomes and job security.

Direct Impacts
The Direct Impacts arise immediately from the conduct of those entities performing the
activity in question. The business operating an activity whose impact is being measured
will have persons on its payroll and will purchase goods and services from suppliers. Both
components are immediately attributable to that business in a full legal and accounting
sense. The expenditures occur on-site, and the items purchased are clearly and
immediately related to the core business. For an airport, the direct impacts would
include the activities of airlines, the airport itself, forwarders, ground handling agents and
other firms whose principal business involves commercial aviation. A direct impact could
include purchases of fluid for deicing aircraft by a handling agency.

Enplaned-Deplaned Traffic
Traffic that gets on board at a given point, and exits the aircraft at the destination point is
and enplaning and deplaning passenger traffic. This is in contract to an on-board
passenger that may have been on the flight when it arrived and will continue on the same
flight as the enplaning passenger. The latter would be called a through, rather than an
enplaning passenger.

B-2
Freedoms of the Air
The Chicago Convention of 1944 provided the framework for todays system of
bilateral Air Service Agreements (ASAs). The convention identified the five freedoms of the
air that defined the terms for future negotiations of air traffic rights.

First freedom rights or the right of innocent passage, is the right to fly over the other
contracting partys territory.

Second freedom rights provide for the airline(s) of one contracting party to stop in the
territory of the other contracting part, for technical reasons only. No traffic is allowed to
enplane or deplane.

Third & Fourth freedom rights allow the carriers of two contracting parties (countries) to
carry traffic between the two countries.

Fifth freedom rights allow the carriers of one contracting party to carry traffic destined for
a point intermediate, or beyond the territory of the other contracting party. In either case,
the service must originate within the territory of the contracting parties.

Sixth freedom traffic is that traffic which originates behind the gateway of one of the
contracting parties, and is destined for the territory of the other contracting party. For
example, hub carriers in Europe often carry traffic from points in the Middle East or Gulf
States to a hub in Europe for the sole purpose of moving that traffic from its origin to
another destination in Europe or North America. The practice is common, often creates
contentiousness, and is not covered by bilateral agreements.

Seventh freedom rights allow the carriers of one contracting party to carry traffic between
two countries without any connection to the home country. Seventh freedom operations
often take place among global all-cargo air carriers and are very infrequent in passenger
operations.

Full-Time Equivalent Position


A quantity of employment corresponding to a full-time position. One full-time employee or
two half-time employees would account for one Full-Time Equivalent position.

Gross Domestic Product


The Gross Domestic Product (GDP) measures the total value of goods and services
produced in a country during a specific period of time. It includes exports and dividends
paid to foreigners, but excludes imports and dividends or interest received from outside
the entity. The GDP will therefore not necessarily agree with the total value of
consumption. The GDP is the most common measure of the level of economic activity
within a particular area.

The size and strength of an economy, as measured by the GDP, is an important


determinant of many sectoral variables throughout an economy, including air traffic. It

B-3
was therefore considered an independent variable for the Air Route Model; a factor that
can vary freely and autonomously, and which will be reflected in the dependent variables
of air traffic. In reality, many economic quantities influence traffic personal income,
wealth, corporate profits, imports, experts, etc. Those variables are so closely associated
with the GDP that the estimation algorithms cannot disentangle their separate influences.

Indirect Impacts
The Indirect Impacts involve the supply chain of the businesses or entities conducting
the primary activity. The airlines at an airport may purchase goods or services, such as
stationery and office supplies, from a local business. The items purchased can be used for
many purposes besides commercial aviation, and would usually occur off-site. The
materials support the primary aviation activity, although they could be used for many
purposes.

Induced Impacts
Many persons and businesses will receive revenues and income from the direct and
indirect expenditures. They will spend a portion of these gains on goods and services. The
recipients of these expenditures will, in turn, spend a portion of them as well. The process
will continue indefinitely, with each successive round of spending smaller than the one
preceding it. The sum of these expenditures will be a multiple of the original injection of
funds from the direct and indirect expenditures. The process through which the successive
impacts ultimately exceed the original stimulus is termed the multiplier effect. For
example, airline employees may spend a portion of their incomes at a furniture store. The
employees of this store will spend some of the additional wages at a restaurant. The
furniture store and the restaurant have no business relationship with the airport, but still
benefit by its presence.

Integrator
The Integrators or integrated cargo carriers include companies such as FedEx, United
Parcel Service and DHL. They perform all steps of the shipping process in-house pickup
and delivery, tracing, marketing/sales development, intercity transportation, insurance,
customs brokerage, etc. This is in contrast to the commercial airlines, which use many
outside specialists to perform each of these functions.

Legacy Carriers
The legacy airlines include American, Continental, Delta, Northwest, U.S. Airways and
United. They were established in the 1930s to carry air mail on government contract.
Until 1978, they operated under the strict control of the Civil Aeronautics Board. They
inherited the cost structure, management practices and employee expectations of a
regulated industry. The legacy airlines operate large hub-and-spoke networks, which offer
one-stop connecting service between a large number of even light-volume city pairs. While
their virtually everywhere-to-everywhere networks offer major revenue advantages, the
operational complexity add to costs. The legacy carriers have encountered challenges
cutting their costs so as to be able to compete with the low cost carriers. The solution has
often been to restructure under the protection of chapter 11 of the bankruptcy code.

B-4
Low Cost Carriers (LCCs)
The Low Cost Carriers (LCCs) include Southwest Airlines, AirTran, Spirit Airlines, Frontier
Airlines and jetBlue. These airlines came into being and grew outside of the protective
umbrella of the Civil Aeronautics Board (CAB), and therefore did not inherit the cost
structure and inefficiencies that were embedded in the LNCs. Southwest began by serving
only intra-Texas markets. Since it operated solely within Texas, it was not subject to
federal jurisdiction. It expanded to interstate routes once the United States deregulated
interstate aviation in 1978. The other LCCs are among many airlines that started
operations after 1978; most others have failed.

While the network and fleet strategies of the different LCCs vary, all share a tight focus on
costs. All operate a simplified network, a fleet of one or two aircraft types, work rules that
permit high productivity and a basic product (limited inflight service, no through ticketing
with other airlines, a small number of fares for a specific trip, etc.). Within these limits,
the services do vary some operate hub and spoke services; others offer business class,
etc.

As air travel becomes an increasingly homogeneous and price-sensitive commodity, the


low fare airlines have grown at the expense of the legacy airlines. The legacy carriers have
been increasingly forced to lower their costs, and emulate many of the attributes of the
LCCs to survive.

Memorandum of Understanding (MOU)


Two nations with an effective Air Service Agreement may wish to make incremental
modifications to the regime. Such changes could include allowing additional capacity,
resolving an ongoing dispute, clarifying any ambiguities or definitions, or clarifying items
that had been left to be agreed in the original negotiations. A total renegotiation of the
agreement could be procedurally difficult for either party, or both nations might be
satisfied with the overall framework. Under such circumstances, the countries would
agree to retain the original agreement but amend it as necessary. The results of the
negotiations would be summarized in a Memorandum of Understanding, Record of
Consultations, Exchange of Notes or similar mechanisms. Although the parties agree to
retain the original agreement, the negotiations can be very complicated and important.

Open Skies
An Open Skies air service agreement creates a very liberal market between the two
signatory nations. It allows any number of airlines from either nation unlimited rights to
fly between any city-pair involving the two countries. The airlines can, moreover, carry
revenue traffic to and from any third countries, subject to appropriate provisions in the
other agreements. An open skies agreement has no restrictions on capacity or code
sharing. Fares proposed by any airline of either country prevail unless both nations
oppose them. The open skies agreement is often viewed as the ideal form of liberalization
because it encourages the full play of market forces.

B-5
Most such agreements do not allow cabotage rights, in which one airline of the signatory
nation could carry wholly domestic traffic of the other country. Some countries such as
the United States view an open skies agreement as a prerequisite for antitrust immunity.
The number of nations embracing the concept of open skies has grown rapidly, but most
recognize such agreements as pure a waypoint on the route to a truly international and
multinational industry.

Origin-Destination Traffic
Traffic that is originating its trip in a given city, and is destined for a second city on the
flight. This is what is referred to as true Origin-Destination traffic (O&D) as opposed to the
traffic on-board the same flight, much of which might be destined for a city beyond the
first stop of the flight, or have originated at a city behind the origin of the flight.

R-Square
An econometric or regression model expresses one series of data in terms of one or more
other series. For example, it may express T.F. Greens traffic to Florida in terms of
average fares for the route, airline capacity, Rhode Island population, average incomes or
other variables. It is very useful to examine the historical values of the traffic, and
compare them to the projections provided by the model. In some years, the model may
over-estimate traffic; in other years, it may predict too few passengers. It will never
provide an exact fit. However, some models provide a closer fit than others.

The r-square measures how well a particular model explains the given series of data. A
value close to one shows that its predictions are very close to the actual values; a value
close to zero indicates that the model does not explain the series very well.

The r-square refers only to the ability of the model to explain the data actually used to
estimate the model. It does not necessarily show that the model is suitable for forecasting.
The r-square is only one measure of the quality of a model. Of much greater importance,
the model should provide valid and intuitively sensible predictions about how changes in
one variable will cause changes in another. A model that suggests that air traffic will grow
if GDP falls and fares rise should be swiftly rejected whatever the value of its r-square.

The r-square is a simple and easily visualized measure that often substitutes for genuine
analysis. In many situations, there is so much random noise that the variables are
inherently difficult to predict. The r-square could be very low even in a fully valid model.
However, it is always possible to develop a totally invalid model with a high r-square.

Regression (or Econometric) Model


A regression model uses a simple linear equation to express one variable in terms of one
or more other variables. It uses various criteria to estimate the terms of the equation to
yield the best fit. For example, a regression model could express passenger traffic from
Boston to Florida in terms of the number of daily nonstop seats and the average fare. In
this example, the number of seats and the fare are considered as being given, and the
traffic is then determined through its relationship to the two variables. The prediction is

B-6
usually imperfect; some random event such as a hurricane in Florida, temporary deep-
discount fares to a competing destination or a recession may cause the model to under- or
over-predict. There are many measures to judge the quality of a model.

If future values of the seat capacity and fare are known, this model will provide forecasts.
The forecasts will only be valid if the relationships between the variables that are
estimated from past data remain valid in the future. For example, a future change in
competitive market fares could change the future Boston-Florida results in some
systematic manner. This problem can be avoided by choosing the correct variables in the
equations.

Yield
Yield is an aggregate average measure of the cost of air travel. It consists of the total
revenues divided by the sum of the miles traveled by every paying passenger. The units of
the measure are dollars per revenue passenger-mile. The measure is sometimes used in
combination with the fare. The fare is the dollars paid per passenger. Fares for a long
distance journey are usually higher than for a shorter trip. However, the fare usually rises
at a lower proportional rate than the increased distance. The yield on a long distance
market is therefore usually lower than for a short distance city-pair, even if the fares are
higher.

B-7
InterVISTAS-ga2 Consulting, Inc.
1800 K Street NW
Suite 1104
Washington, DC 20006
Voice: 202-457-0212
Fax: 202-833-3183
web: www.intervistas.us

Вам также может понравиться